_ registry / mcp http-sse · checked 15m ago

aceround

https://api.aceround.app

Registry code: 546a50bcdfde51c9

api record

AceRound: review interview and exam reports, manage resumes, practice mock interviews via OAuth

from a public catalogue that lists it, not from the operator

endpoint
https://api.aceround.app/agent/mcp
protocol
http-sse ·2025-06-18
authentication
none observed
public key
none — nobody has proven they own this listing · is it yours? claim it
karma
0 · newcomer
_ is it live, free and safe measured by this hub
Is aceround live?
Yes — it answered the hub's last check (checked 15m ago). It answered 100% of checks over the last 30 days.
Is aceround free to use?
Not measured yet.
What tools does aceround have?
129 tools: get_campus_job_filters, list_interviews, delete_hot_word_library, update_exam_customization, get_sms_relay_status, rename_interview, submit_job_apply_sms_code, save_custom_prompt, ….
Is aceround safe to connect?
The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.
reachable
live
uptime, 30 days
100%

90 days 100%· all time 100%

latency
677ms

last good check

priced tools
0

of 129 tools

_ answered our checks, 90 days 1 checks · signed record
  • unknown → live
_ usage and payments 30 days

Calls placed through this hub's router, from its own receipts. Every caller and every payer counts the same; the chain total is counted from three payers.

accounts
0

through this hub

calls served
0

successful

paid through this hub
0 USDC

what callers paid

_ what it can do 129 tools
129 never probed 0 of 129 classified

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

  • get_campus_job_filters unknown never probed

    Filter values for the job board (employment types, job functions, locations, companies) and its size.

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • list_interviews unknown never probed

    List the user's interview sessions (live and mock), newest first, 20 per page. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "page": {
          "type": "integer",
          "default": 1,
          "maximum": 100,
          "minimum": 1
        },
        "type": {
          "enum": [
            "INTERVIEW",
            "MOCK_INTERVIEW"
          ],
          "type": "string",
          "description": "INTERVIEW = live interviews, MOCK_INTERVIEW = mock interviews; omit for both"
        }
      },
      "additionalProperties": false
    }
    arguments 20 lines
  • delete_hot_word_library unknown never probed

    Delete a custom hotword library; roles bound to it are unbound. The ID comes from list_my_hot_words. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • update_exam_customization unknown never probed

    Update the screenshot-answer customization used in exams and interviews (synced; the desktop client picks it up on next launch). Pass only what changes. Current values via get_settings; save a full custom prompt with save_custom_prompt. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "codeMode": {
          "enum": [
            "auto",
            "acm",
            "core"
          ],
          "type": "string",
          "description": "Code mode: auto / acm (full program with I/O) / core (only the required function)"
        },
        "codeQuality": {
          "enum": [
            "standard",
            "readable",
            "concise",
            "performance"
          ],
          "type": "string",
          "description": "Code style: standard / readable / concise / performance"
        },
        "quickPrompts": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 60
          },
          "maxItems": 8,
          "description": "Quick prompts shown in the answer overlay (replaces the list, up to 8)"
        },
        "answerLanguage": {
          "anyOf": [
            {
              "enum": [
                "zh-CN",
                "yue-CN",
                "zh-HK",
                "en-US",
                "ja-JP",
                "es-MX",
                "pt-BR",
                "de-DE",
                "fr-FR",
                "ko-KR",
                "vi-VN",
                "ru-RU",
                "ar-SA"
              ],
              "type": "string"
            },
            {
              "type": "string",
              "const": "auto"
            }
          ],
          "description": "Explanation language; auto follows the UI language in exams and the interview language in interviews"
        },
        "customRequirements": {
          "type": "string",
          "maxLength": 20000,
          "description": "Extra requirements"
        },
        "programmingLanguage": {
          "type": "string",
          "maxLength": 80,
          "description": "Programming language, e.g. Python, Java, C++; auto follows the problem"
        }
      },
      "additionalProperties": false
    }
    arguments 71 lines
  • get_sms_relay_status unknown never probed

    Show SMS code forwarding status: whether the Android SMS forwarder is connected, when the iPhone shortcut last pushed, and whether the forwarder app needs an update. The apply assistant uses it to read sign-in codes. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • rename_interview unknown never probed

    Rename an interview record (live or mock). (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId",
        "title"
      ],
      "properties": {
        "title": {
          "type": "string",
          "maxLength": 60,
          "minLength": 1
        },
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • submit_job_apply_sms_code unknown never probed

    Give the SMS code the user received to the apply AI, the same as typing it into the code box at the top of the app's main window: if the AI is still waiting, the code goes straight to that wait and the turn continues; if the wait has ended, it is sent to that conversation as a message. taskId can be omitted when only one application is waiting; pass it when several are (see get_job_apply_client_status). Says so if nothing is waiting for a code. Replies with the employer / role it went to. Requires the AceRound Auto Apply desktop app on the user’s computer to be signed in and open (running in the background after closing the window also works); if the app is offline or outdated, the tool says so. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "code"
      ],
      "properties": {
        "code": {
          "type": "string",
          "maxLength": 20,
          "minLength": 1,
          "description": "The SMS code"
        },
        "taskId": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Conversation id (taskId), from get_job_apply_client_status"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • save_custom_prompt unknown never probed

    Save the full custom prompt from the prompt studio: when non-empty it replaces the built-in instructions for interview answers or exam answers; an empty string goes back to the built-in style plus extra requirements. baseline records which official text this prompt was derived from for future official-update merges: if you started from a text in get_prompt_presets pass that one (official / current / style:<value>), after a sync rewrite pass official, for a prompt written from scratch pass none, and omit it for small edits to keep the existing record. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "scene",
        "prompt"
      ],
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "description": "Interview language this prompt is for; defaults to the saved one"
        },
        "scene": {
          "enum": [
            "interview",
            "exam"
          ],
          "type": "string",
          "description": "interview answers / exam answers"
        },
        "prompt": {
          "type": "string",
          "maxLength": 20000,
          "description": "Full prompt; empty string = stop customizing"
        },
        "baseline": {
          "enum": [
            "official",
            "current",
            "none",
            "style:keypoints",
            "style:answerOnly",
            "style:natural",
            "style:structured",
            "style:standard",
            "style:readable",
            "style:concise",
            "style:performance"
          ],
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 58 lines
  • get_product_overview unknown never probed

    AceRound's product overview: what each product does, supported interview and assessment platforms, downloads, and links to pricing and help pages. Read it before answering any question about AceRound.

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • list_plans unknown never probed

    List the plans on sale, with prices and what each one includes.

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_purchase_link unknown never probed

    Create a purchase link for a plan for the user to open in a browser: after signing in to AceRound, checkout for that plan opens automatically, priced for the user's country. Product ids come from list_plans.

    mcp-tool

    {
      "type": "object",
      "required": [
        "productId"
      ],
      "properties": {
        "productId": {
          "type": "string",
          "maxLength": 64,
          "description": "A product id from list_plans"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • search_campus_jobs unknown never probed

    Search open roles on AceRound's job board and return the first page of results, one role per line. Apply links and further pages require signing in on the website.

    mcp-tool

    {
      "type": "object",
      "properties": {
        "city": {
          "type": "string",
          "maxLength": 32,
          "description": "Work location, from the city line of get_campus_job_filters"
        },
        "sort": {
          "enum": [
            "hot",
            "recent",
            "published"
          ],
          "type": "string",
          "default": "hot",
          "description": "hot = most popular, recent = most recently found, published = most recently posted"
        },
        "company": {
          "type": "string",
          "maxLength": 64,
          "description": "Company name"
        },
        "keyword": {
          "type": "string",
          "maxLength": 64,
          "description": "Keyword in the job title or description, such as \"backend\" or \"product manager\""
        },
        "jobFunction": {
          "type": "string",
          "maxLength": 32,
          "description": "Job function, one of the jobFunction values from get_campus_job_filters"
        },
        "employmentType": {
          "type": "string",
          "maxLength": 16,
          "description": "Employment type, one of the employmentType values from get_campus_job_filters"
        }
      },
      "additionalProperties": false
    }
    arguments 41 lines
  • get_account unknown never probed

    The signed-in account: remaining minutes, pass level and expiry, mock interview and prediction credits, Auto Apply credits, and the payment status of the 10 most recent orders. Use it to confirm a purchase made through a purchase link went through. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_interview_report unknown never probed

    Read the review report of one interview: summary, scores, strengths and weaknesses, suggestions, and each question with the answer, score, evaluation, and suggestion. Interview ids come from list_interviews. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • get_interview_transcript unknown never probed

    Read the full transcript of one interview (interviewer and user turns in order, 80 per call; page with offset). (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "offset": {
          "type": "integer",
          "default": 0,
          "minimum": 0
        },
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 19 lines
  • list_resumes unknown never probed

    List the user's resumes, one per line. A resume marked sample is the built-in example, not the user's own background. (Requires the user's sign-in with scope resumes:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_resume unknown never probed

    Read the text of one resume. Resume ids come from list_resumes. (Requires the user's sign-in with scope resumes:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "resumeId"
      ],
      "properties": {
        "resumeId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • list_job_applications unknown never probed

    List the user's Auto Apply applications: overall counts, then each application's id, company, role, status, and progress (50 per page). Tools that act on an application take the id from here. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "page": {
          "type": "integer",
          "default": 1,
          "maximum": 100,
          "minimum": 1
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • get_login_devices unknown never probed

    Show how many devices signed in to this account in the last 7 days and whether more than one computer (web or desktop app) was used, to tell whether the account is being shared. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • redeem_credits unknown never probed

    Convert all account credits into interview minutes and mock interviews: every 200 credits become 60 minutes and 2 mock interviews. All credits are converted at once and any remainder below 200 is cleared. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_gift_code_info unknown never probed

    Look up a gift code: what it includes, whether it has expired or run out, and whether this account can redeem it. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "code"
      ],
      "properties": {
        "code": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Gift code"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • redeem_gift_code unknown never probed

    Redeem a gift code; its contents are added to this account immediately. Each code can be redeemed once per account. Use get_gift_code_info to preview it. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "code"
      ],
      "properties": {
        "code": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Gift code"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • list_pending_subscription_cards unknown never probed

    List the user's purchased passes that have not been activated yet (bought with delayed activation), and whether each can be refunded and for how much. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • activate_subscription_card unknown never probed

    Activate an unactivated pass: it starts now, or right after the current pass if one is active. An activated pass can no longer be refunded. Card ids come from list_pending_subscription_cards. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "cardId"
      ],
      "properties": {
        "cardId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • refund_subscription_card unknown never probed

    Refund an unactivated pass to the original payment method; the pass is voided. Only passes showing a refundable amount in list_pending_subscription_cards qualify. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "cardId"
      ],
      "properties": {
        "cardId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • list_agent_grants unknown never probed

    List the AI assistants this account has authorized: name, site, scopes, when access was granted, and when it was last used. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • revoke_agent_grant unknown never probed

    Revoke an AI assistant's access; its tokens stop working immediately. Revoking the current assistant's own grant means the user must authorize it again. Grant ids come from list_agent_grants. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "grantId"
      ],
      "properties": {
        "grantId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • check_order_status unknown never probed

    Check whether an order has been paid and fulfilled. Use it after the user pays at checkout; PAID means the purchase is on the account. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "orderNumber"
      ],
      "properties": {
        "orderNumber": {
          "type": "string",
          "maxLength": 128,
          "minLength": 1,
          "description": "Order number returned when the order was created, or from the recent orders in get_account"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • create_payment_order unknown never probed

    Create an order for a plan and return a Stripe Checkout link for the user to open and pay. The price is the local price for the country this request comes from and is shown in the result. After the user pays, use check_order_status to confirm. Product ids come from list_plans. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "productId"
      ],
      "properties": {
        "productId": {
          "type": "string",
          "maxLength": 64,
          "description": "A product id from list_plans"
        },
        "deferActivation": {
          "type": "boolean",
          "description": "Passes only: true = keep the pass unactivated after payment until the user activates it (activate_subscription_card)"
        }
      },
      "additionalProperties": false
    }
    arguments 18 lines
  • buy_auto_apply_credits unknown never probed

    Create a Stripe Checkout link for Auto Apply credits: plan "pack" = 50 applications, "unlimited" = 30 days of unlimited applications. The user opens the link and pays; once paid, the credits show up in get_account. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "plan"
      ],
      "properties": {
        "plan": {
          "enum": [
            "pack",
            "unlimited"
          ],
          "type": "string",
          "description": "pack = 50 applications; unlimited = 30 days unlimited"
        }
      },
      "additionalProperties": false
    }
    arguments 17 lines
  • submit_feedback unknown never probed

    Submit feedback to AceRound on the user's behalf (a bug, a suggestion, or anything else), optionally with screenshots. Follow replies with list_my_feedbacks / get_my_feedback. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "message"
      ],
      "properties": {
        "type": {
          "enum": [
            "BUG",
            "SUGGESTION",
            "OTHER"
          ],
          "type": "string",
          "default": "OTHER",
          "description": "BUG = something is broken, SUGGESTION = idea, OTHER = anything else"
        },
        "title": {
          "type": "string",
          "maxLength": 60,
          "minLength": 2,
          "description": "Title; defaults to the start of the message"
        },
        "message": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 4,
          "description": "Feedback text"
        },
        "notifyEmail": {
          "type": "string",
          "format": "email",
          "maxLength": 200,
          "description": "Email to notify when there is a reply; omit for in-app notices only"
        },
        "screenshots": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "fileName",
              "contentBase64"
            ],
            "properties": {
              "fileName": {
                "type": "string",
                "maxLength": 200,
                "minLength": 1
              },
              "contentBase64": {
                "type": "string",
                "minLength": 1,
                "description": "Image file content as base64 (PNG / JPG / WEBP)"
              }
            },
            "additionalProperties": false
          },
          "default": [],
          "maxItems": 6,
          "description": "Screenshots, up to 6"
        }
      },
      "additionalProperties": false
    }
    arguments 63 lines
  • list_my_feedbacks unknown never probed

    List the user's feedback, one per line, most recently updated first, 20 per page: status, number of replies, unread replies, and the latest message. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "page": {
          "type": "integer",
          "default": 1,
          "maximum": 100,
          "minimum": 1
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • get_my_feedback unknown never probed

    Read the full thread of one feedback item (every message between the user and support); this marks its replies as read. Ids come from list_my_feedbacks. (Requires the user's sign-in with scope account:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "feedbackId"
      ],
      "properties": {
        "feedbackId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • reply_to_my_feedback unknown never probed

    Add a message (optionally with screenshots) to one of the user's feedback threads; support will follow up. Ids come from list_my_feedbacks. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "feedbackId",
        "content"
      ],
      "properties": {
        "content": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1
        },
        "feedbackId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        },
        "screenshots": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "fileName",
              "contentBase64"
            ],
            "properties": {
              "fileName": {
                "type": "string",
                "maxLength": 200,
                "minLength": 1
              },
              "contentBase64": {
                "type": "string",
                "minLength": 1,
                "description": "Image file content as base64 (PNG / JPG / WEBP)"
              }
            },
            "additionalProperties": false
          },
          "default": [],
          "maxItems": 6,
          "description": "Screenshots, up to 6"
        }
      },
      "additionalProperties": false
    }
    arguments 46 lines
  • update_feedback_notify_email unknown never probed

    Change the email notified when a feedback item gets a reply; pass null to stop emails and keep in-app notices only. Ids come from list_my_feedbacks. (Requires the user's sign-in with scope account:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "feedbackId",
        "notifyEmail"
      ],
      "properties": {
        "feedbackId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        },
        "notifyEmail": {
          "anyOf": [
            {
              "type": "string",
              "format": "email",
              "maxLength": 200
            },
            {
              "type": "null"
            }
          ]
        }
      },
      "additionalProperties": false
    }
    arguments 27 lines
  • upload_resume unknown never probed

    Upload a resume file (PDF / JPG / PNG / WEBP). The server reads its text and saves it as a new resume for interviews and Auto Apply. The file content is passed as base64. Returns the new resume id; read it back with get_resume. (Requires the user's sign-in with scope resumes:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "fileName",
        "fileBase64"
      ],
      "properties": {
        "title": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Resume title; defaults to the file name"
        },
        "fileName": {
          "type": "string",
          "maxLength": 200,
          "minLength": 1,
          "description": "File name, such as \"resume.pdf\""
        },
        "fileBase64": {
          "type": "string",
          "minLength": 1,
          "description": "File content as base64"
        }
      },
      "additionalProperties": false
    }
    arguments 27 lines
  • create_resume_from_text unknown never probed

    Create a new resume from text (for example a revised version of the user's resume) for interviews and Auto Apply. No file is produced and existing resumes are left as they are. Returns the new resume id. (Requires the user's sign-in with scope resumes:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "title",
        "content"
      ],
      "properties": {
        "title": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Resume title"
        },
        "content": {
          "type": "string",
          "maxLength": 50000,
          "minLength": 1,
          "description": "Full resume text"
        }
      },
      "additionalProperties": false
    }
    arguments 22 lines
  • delete_resume unknown never probed

    Delete one of the user's resumes. Resume ids come from list_resumes. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope resumes:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "resumeId"
      ],
      "properties": {
        "resumeId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • delete_interview unknown never probed

    Delete an interview record, including its report and transcript. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • retry_interview_report unknown never probed

    Regenerate an interview report that failed to generate (when get_interview_report says it failed). Generation runs in the background for a minute or two; read it afterwards with get_interview_report. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • generate_interview_experience unknown never probed

    Write up an interview experience post (the questions the interviewer asked) from a live interview record, save it to the record, and return it; an existing write-up is replaced. Only live interviews of at least 5 minutes qualify. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • export_interview_markdown unknown never probed

    Turn a full interview record (transcript and answer-assistant solutions) into tidy Markdown. Returns its length, outline, opening, and a download link for the full text. Takes a few dozen seconds. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • list_answer_style_sources unknown never probed

    List interview records that can be used to learn the user's answer style (sessions with enough of the user's own answers), newest first. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "limit": {
          "type": "integer",
          "default": 10,
          "maximum": 30,
          "minimum": 1
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • learn_answer_style unknown never probed

    Learn the user's own answer style from one interview record and return a style instruction that can be used as the extra answer requirement when starting an interview. Interview ids come from list_answer_style_sources. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "interviewId"
      ],
      "properties": {
        "interviewId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Interview id from list_interviews"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • get_voiceprint_status unknown never probed

    Show whether the user has enrolled a voiceprint (used in phone-interview mode to tell the user's voice from the interviewer's), when, and whether the voiceprint service is available. Enrollment is done by recording in the audio settings when starting an interview. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • list_exams unknown never probed

    List the user's assessment records (screenshot-solving sessions in the desktop app), one per line, most recently updated first, 20 per page. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "page": {
          "type": "integer",
          "default": 1,
          "maximum": 100,
          "minimum": 1
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • get_exam unknown never probed

    Read one assessment record: summary plus one line per question (start of the question, answer length). Pass question (1-based) to include that question and its answer in full. Use export_exam_markdown for the whole record. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "examId"
      ],
      "properties": {
        "examId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Assessment id from list_exams"
        },
        "question": {
          "type": "integer",
          "minimum": 1,
          "description": "Question number, starting at 1"
        }
      },
      "additionalProperties": false
    }
    arguments 20 lines
  • update_exam unknown never probed

    Rename an assessment record, or change or clear the role it is linked to. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "examId"
      ],
      "properties": {
        "title": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1
        },
        "examId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Assessment id from list_exams"
        },
        "goalId": {
          "anyOf": [
            {
              "type": "string",
              "maxLength": 64,
              "minLength": 1
            },
            {
              "type": "null"
            }
          ],
          "description": "The user's role (goal) id; null clears the link, omit to keep it"
        }
      },
      "additionalProperties": false
    }
    arguments 33 lines
  • delete_exam unknown never probed

    Delete an assessment record with all its questions and answers. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "examId"
      ],
      "properties": {
        "examId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Assessment id from list_exams"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • export_exam_markdown unknown never probed

    Turn a whole assessment record (every question and answer) into tidy Markdown with code kept as is. Returns its length, outline, opening, and a download link for the full text. Takes a few dozen seconds. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "examId"
      ],
      "properties": {
        "examId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Assessment id from list_exams"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • start_exam unknown never probed

    Start an assessment on the user's computer, exactly like clicking "Start assessment" on the assessment page of the desktop app: starts with the given answer mode and role, opens the assessment overlay and tucks away the main window. Requires the desktop app to be running and signed in on the user's computer; if it is offline, an older version, short on balance, or an interview or assessment is already in progress, the reason is returned and nothing starts. Starting is free; each screenshot solved afterwards deducts the user's minutes by the chosen mode. Once started, the user uses the overlay on the assessment page as usual; solve_screen_question can also capture and solve. The assessment record is created on the first solved screenshot and can then be seen with list_exams. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "mode": {
          "enum": [
            "instant",
            "quick",
            "programming"
          ],
          "type": "string",
          "description": "Answer mode: instant = Instant mode(2 minutes per screenshot); quick = Quick mode(2 minutes per screenshot); programming = Coding mode(5 minutes per question). Members are not charged minutes. Omit to keep the mode currently selected on the assessment page of the app"
        },
        "goalId": {
          "anyOf": [
            {
              "type": "string",
              "maxLength": 64,
              "minLength": 1
            },
            {
              "type": "null"
            }
          ],
          "description": "Role (target) ID to link, from list_interview_targets; null = no role. Use either this or jobName; omit both to keep the role currently filled in on the assessment page of the app"
        },
        "jobName": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Name of a new role, when the user has not saved it yet; it is created on start (same as typing a role on the assessment page)"
        },
        "companyName": {
          "type": "string",
          "maxLength": 100,
          "description": "Company of the new role, given with jobName"
        }
      },
      "additionalProperties": false
    }
    arguments 39 lines
  • end_exam unknown never probed

    Mark an in-progress assessment record as ended and give it a title automatically. This only changes the record; it does not close the assessment window on the user's computer. Assessments can only be started and solved in the desktop app. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "examId"
      ],
      "properties": {
        "examId": {
          "type": "string",
          "maxLength": 64,
          "minLength": 1,
          "description": "Assessment id from list_exams"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • list_interview_targets unknown never probed

    List the user's saved interview targets (company + role). Starting a live or mock interview needs one of these IDs. One per line, newest first. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • create_interview_target unknown never probed

    Create an interview target (company + role, optionally a job description) and return its ID for starting an interview. Same as "New target" on the website. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "name"
      ],
      "properties": {
        "name": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Role title"
        },
        "companyName": {
          "type": "string",
          "maxLength": 100,
          "description": "Company name"
        },
        "description": {
          "type": "string",
          "maxLength": 10000,
          "description": "Job description, used as context when answering"
        }
      },
      "additionalProperties": false
    }
    arguments 25 lines
  • start_interview unknown never probed

    Start a live or mock interview for the user and return the interview page link for the user to open in a browser (web only; it does not launch the desktop app; the user must be signed in to the same account in that browser). Live: the user shares the meeting tab with audio as prompted and, after clicking "Start interview", minutes are deducted by time online. Mock: one mock interview credit is deducted on creation. Only one session of each type can be in progress. Advanced settings not given (answer style, custom instructions, translation, answer recording, web search, knowledge bases) follow the user's saved interview preferences. Target IDs come from list_interview_targets, resume IDs from list_resumes. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "targetId",
        "resumeId"
      ],
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "description": "Interview language (selects the speech recognition model); omit to use the user's preference"
        },
        "mode": {
          "enum": [
            "online",
            "phone"
          ],
          "type": "string",
          "default": "online",
          "description": "Live only: online = video meeting (captures meeting audio), phone = phone interview (speakerphone picked up by the microphone; needs an enrolled voiceprint; Chinese / Cantonese / English only)"
        },
        "type": {
          "enum": [
            "INTERVIEW",
            "MOCK_INTERVIEW"
          ],
          "type": "string",
          "default": "INTERVIEW",
          "description": "INTERVIEW = live interview, MOCK_INTERVIEW = mock interview"
        },
        "focus": {
          "type": "string",
          "maxLength": 500,
          "description": "Mock only: what this session should focus on"
        },
        "round": {
          "enum": [
            "FIRST",
            "SECOND",
            "THIRD",
            "HR",
            "OTHER"
          ],
          "type": "string",
          "description": "Live only: which round this is for the target"
        },
        "resumeId": {
          "type": "string",
          "minLength": 1,
          "description": "Resume ID"
        },
        "targetId": {
          "type": "string",
          "minLength": 1,
          "description": "Target ID"
        },
        "skipIntro": {
          "type": "boolean",
          "description": "Mock only: skip the self-introduction"
        },
        "assistMode": {
          "type": "boolean",
          "default": false,
          "description": "Mock only: turn on AI assist (a suggested answer for each question)"
        },
        "difficulty": {
          "enum": [
            "FRESH_GRADUATE",
            "1_3_YEARS",
            "3_5_YEARS",
            "5_10_YEARS",
            "10_YEARS"
          ],
          "type": "string",
          "default": "FRESH_GRADUATE",
          "description": "Mock only: seniority level"
        },
        "extraLangs": {
          "type": "array",
          "items": {
            "$ref": "#/properties/lang"
          },
          "maxItems": 3,
          "description": "Live only: other languages the user may switch to during the interview; omit to use the user's preference"
        },
        "basicsRatio": {
          "type": "integer",
          "maximum": 100,
          "minimum": 0,
          "description": "Mock only: share of fundamentals questions, 0–100; omit to let the AI decide"
        },
        "questionCount": {
          "type": "integer",
          "maximum": 24,
          "minimum": 6,
          "description": "Mock only: number of questions; omit to let the AI decide"
        },
        "knowledgeBaseIds": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "description": "Knowledge base IDs; for live interviews omit to use the user's preference; for mock interviews they apply only with AI assist on"
        }
      },
      "additionalProperties": false
    }
    arguments 121 lines
  • get_live_session_status unknown never probed

    Show the user's live interview, mock interview and assessment in progress: session ID, target, start time, expiry, whether the page is online, which client handles it (desktop app / web), and page state (started, AI answering, language, auto-answer, one card per question, etc.). (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • end_live_interview unknown never probed

    End the user's live or mock interview in progress, same as "End interview" on the page: settles the session's minutes (live interviews refund time billed after the last exchange) and starts the report. The page closes itself within seconds. This cannot be undone. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "type": {
          "enum": [
            "INTERVIEW",
            "MOCK_INTERVIEW"
          ],
          "type": "string",
          "default": "INTERVIEW",
          "description": "INTERVIEW = live interview, MOCK_INTERVIEW = mock interview"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • set_live_interview_options unknown never probed

    Change settings of the live interview in progress: recognition language (only within the set chosen at the start; the page reconnects once), auto-answer (when off, answers only on "Answer now"), and one card per question (follow-ups on the same question share one card). Auto-answer and one-card-per-question are supported only by the desktop app overlay. Pass only what should change. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "description": "Recognition language"
        },
        "turnCards": {
          "type": "boolean",
          "description": "One card per question"
        },
        "autoAnswer": {
          "type": "boolean",
          "description": "Auto-answer"
        }
      },
      "additionalProperties": false
    }
    arguments 33 lines
  • ask_live_interview unknown never probed

    Ask a question in the live interview in progress on the user's behalf: the text is treated as the interviewer's question and answered on the interview page; if the AI is answering, that answer is interrupted first. Uses the same path as "Voice ask" in the desktop app and requires the user to have clicked "Start interview". Billed by time online. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "question"
      ],
      "properties": {
        "question": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1,
          "description": "The question"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • answer_live_interview_now unknown never probed

    Press "Answer now" in the live interview in progress: answers the interviewer's latest unanswered speech (at least 6 characters) immediately, interrupting any answer in progress. Use it to trigger answers when auto-answer is off. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • interrupt_live_interview_answer unknown never probed

    Stop the answer the AI is currently writing in the live interview (what is already written stays). Only works while an answer is being written. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • answer_mock_interview unknown never probed

    Submit the user's answer (text) to the current question in the mock interview in progress, same as editing and clicking "Send" on the page; the AI interviewer then asks the next question or ends the interview. Not possible while the interviewer is still asking. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "answer"
      ],
      "properties": {
        "answer": {
          "type": "string",
          "maxLength": 5000,
          "minLength": 1,
          "description": "The answer"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • set_mock_interview_assist unknown never probed

    Turn AI assist (a suggested answer for each question) on or off in the mock interview in progress, same as the switch on the page. Requires the user to have this session open in the browser or desktop app and connected; the page carries it out exactly as if the user clicked the matching button. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "enabled"
      ],
      "properties": {
        "enabled": {
          "type": "boolean"
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • solve_screen_question unknown never probed

    Have the user's running page capture the user's screen now and solve the question, same as the "Solve from screenshot" button / hotkey; charged by the current mode and page count, with the answer shown on the user's page. The AI assistant does not (and cannot) supply the image. Runs in the desktop assessment overlay when the user is in an assessment (it must have started); otherwise in screenshot solving on the live / mock interview page (the web page must be sharing the screen). Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • follow_up_screen_question unknown never probed

    Ask a follow-up about the current screenshot question on the user's page (e.g. "another approach", "explain line 3"), same as typing a follow-up on the page; not possible while an answer is being written. Assessment follow-ups are billed by mode. Runs in the desktop assessment overlay when the user is in an assessment (it must have started); otherwise in screenshot solving on the live / mock interview page (the web page must be sharing the screen). Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "message"
      ],
      "properties": {
        "message": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1,
          "description": "The follow-up"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • answer_exam_now unknown never probed

    Press "Answer now" in the desktop assessment overlay: stops any checking still running and writes the answer from what is already known (it does not cancel; the question still completes). Only works while an assessment question is being answered. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • start_question_prediction unknown never probed

    Predict the questions most likely to come up for one of the user’s target roles, with answer outlines drawn from their resume. Generation takes about 30–90 seconds; the record is saved only when it succeeds. Uses 1 prediction credit on success (PRO is not charged; nothing is charged on failure). If it is still running when this call returns, check list_question_predictions a little later: a new record means it finished; no new record after a few minutes means it failed and nothing was charged. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "goalId"
      ],
      "properties": {
        "lang": {
          "type": "string",
          "maxLength": 40,
          "description": "Output language name, e.g. English, Simplified Chinese; defaults to the user interface language"
        },
        "count": {
          "type": "integer",
          "default": 7,
          "maximum": 10,
          "minimum": 4,
          "description": "Number of questions"
        },
        "goalId": {
          "type": "string",
          "minLength": 1,
          "description": "Target role ID"
        },
        "resumeId": {
          "type": "string",
          "description": "Resume ID from list_resumes; omit to base it on the role only"
        },
        "experience": {
          "type": "string",
          "maxLength": 40,
          "description": "Experience level: Entry-level / 1-3 years / 3-5 years / 5-10 years / 10+ years"
        }
      },
      "additionalProperties": false
    }
    arguments 35 lines
  • list_question_predictions unknown never probed

    List the user’s question predictions: record ID, role, company, created time, status (generating / done / failed) and number of questions. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_question_prediction unknown never probed

    Read one question prediction. The record ID comes from list_question_predictions or start_question_prediction. Returns the status and an overview with one question per line; pass indexes to read the why-asked, answer outline and material for those questions. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Prediction ID"
        },
        "indexes": {
          "type": "array",
          "items": {
            "type": "integer",
            "minimum": 1
          },
          "maxItems": 20,
          "description": "Question numbers to read in full (1-based)"
        }
      },
      "additionalProperties": false
    }
    arguments 23 lines
  • delete_question_prediction unknown never probed

    Delete a question prediction. The record ID comes from list_question_predictions. Used credits are not returned. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Prediction ID"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • save_question_prediction_to_knowledge_base unknown never probed

    Save all questions and answer outlines of a prediction as a new interview knowledge base the interview assistant can use live. The record ID comes from list_question_predictions; returns the new knowledge base ID. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Prediction ID"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • list_knowledge_bases unknown never probed

    List the user’s interview knowledge bases: ID, name, source (custom / file / ai), number of Q&A entries and text length. Knowledge bases are attached to an interview so the assistant’s answers fit the user’s background. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_knowledge_base unknown never probed

    Read one interview knowledge base: name, the start of its text with total length, and an overview of its Q&A entries (one per line). The ID comes from list_knowledge_bases. Pass contentOffset to read long text 4000 characters at a time, or entryIds for full Q&A entries. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Knowledge base ID"
        },
        "entryIds": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "maxItems": 30,
          "description": "Entry IDs to read in full"
        },
        "contentOffset": {
          "type": "integer",
          "minimum": 0,
          "description": "Character offset into the text"
        }
      },
      "additionalProperties": false
    }
    arguments 27 lines
  • create_knowledge_base unknown never probed

    Create an interview knowledge base, optionally with free text (project notes, experience). Returns its ID; add Q&A with add_knowledge_base_entry or import_knowledge_base_entries. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "name"
      ],
      "properties": {
        "name": {
          "type": "string",
          "maxLength": 50,
          "minLength": 1
        },
        "content": {
          "type": "string",
          "maxLength": 90000,
          "description": "Free text"
        }
      },
      "additionalProperties": false
    }
    arguments 19 lines
  • update_knowledge_base unknown never probed

    Rename an interview knowledge base or replace its text; fields you omit stay unchanged. Passing content replaces the whole text. The ID comes from list_knowledge_bases. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Knowledge base ID"
        },
        "name": {
          "type": "string",
          "maxLength": 50,
          "minLength": 1
        },
        "content": {
          "type": "string",
          "maxLength": 90000
        }
      },
      "additionalProperties": false
    }
    arguments 23 lines
  • add_knowledge_base_entry unknown never probed

    Add one Q&A entry (question + reference answer) to an interview knowledge base. The ID comes from list_knowledge_bases; returns the entry ID. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "materialId",
        "question",
        "answer"
      ],
      "properties": {
        "answer": {
          "type": "string",
          "maxLength": 8000,
          "minLength": 1
        },
        "question": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1
        },
        "materialId": {
          "type": "string",
          "minLength": 1,
          "description": "Knowledge base ID"
        }
      },
      "additionalProperties": false
    }
    arguments 26 lines
  • update_knowledge_base_entry unknown never probed

    Edit one Q&A entry; fields you omit stay unchanged. Entry IDs come from get_knowledge_base. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "materialId",
        "entryId"
      ],
      "properties": {
        "answer": {
          "type": "string",
          "maxLength": 8000,
          "minLength": 1
        },
        "entryId": {
          "type": "string",
          "minLength": 1,
          "description": "Entry ID"
        },
        "question": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1
        },
        "materialId": {
          "type": "string",
          "minLength": 1,
          "description": "Knowledge base ID"
        }
      },
      "additionalProperties": false
    }
    arguments 30 lines
  • import_knowledge_base_entries unknown never probed

    Bulk-import Q&A entries (1–200 per call, appended) into an interview knowledge base. The ID comes from list_knowledge_bases; the batch is rejected if it would exceed the base’s capacity. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "materialId",
        "entries"
      ],
      "properties": {
        "entries": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "question",
              "answer"
            ],
            "properties": {
              "answer": {
                "type": "string",
                "maxLength": 8000,
                "minLength": 1
              },
              "question": {
                "type": "string",
                "maxLength": 2000,
                "minLength": 1
              }
            },
            "additionalProperties": false
          },
          "maxItems": 200,
          "minItems": 1
        },
        "materialId": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 38 lines
  • upload_knowledge_base_file unknown never probed

    Create an interview knowledge base from a file: its text becomes the knowledge base text; returns the new ID and text length. Pass the file content as base64. Supported: .pdf .doc .docx .txt .md .jpg .jpeg .png .webp, up to 10MB. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "fileName",
        "fileBase64"
      ],
      "properties": {
        "fileName": {
          "type": "string",
          "maxLength": 255,
          "minLength": 1,
          "description": "File name with extension"
        },
        "fileBase64": {
          "type": "string",
          "minLength": 1,
          "description": "File content as base64"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • generate_knowledge_base_entries unknown never probed

    Have AI predict interview questions from the user’s resume and/or target role and append them with reference answers to a knowledge base (8 by default, up to 15). Knowledge base ID from list_knowledge_bases, resume ID from list_resumes; give at least one of resume or role. Returns how many entries were added. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "materialId"
      ],
      "properties": {
        "lang": {
          "type": "string"
        },
        "count": {
          "type": "integer",
          "maximum": 15,
          "minimum": 1
        },
        "goalId": {
          "type": "string"
        },
        "resumeId": {
          "type": "string"
        },
        "materialId": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 26 lines
  • delete_knowledge_base unknown never probed

    Delete an interview knowledge base with all its entries. The ID comes from list_knowledge_bases. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Knowledge base ID"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • delete_knowledge_base_entry unknown never probed

    Delete one Q&A entry. Entry IDs come from get_knowledge_base. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Entry ID"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • list_official_knowledge_bases unknown never probed

    List the official knowledge bases: recent industry knowledge missing from model training data, kept up to date. They can be attached to an interview alongside the user’s own knowledge bases without importing. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_official_knowledge_base unknown never probed

    Read an official knowledge base: description and an overview of its entries (one per line); pass indexes for the full entries with sources. The ID comes from list_official_knowledge_bases. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Official knowledge base ID"
        },
        "indexes": {
          "type": "array",
          "items": {
            "type": "integer",
            "minimum": 1
          },
          "maxItems": 30,
          "description": "Entry numbers to read in full (1-based)"
        }
      },
      "additionalProperties": false
    }
    arguments 23 lines
  • list_hot_word_libraries unknown never probed

    List the hotword libraries that can be bound to a target role: official libraries and the user’s own (isOwn=true), with IDs and names. Hotwords help speech recognition get proper nouns right. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • list_my_hot_words unknown never probed

    List the user’s own hotword libraries with all their words. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • create_hot_word_library unknown never probed

    Create a custom hotword library (name + words; up to 200 words per library and 20 libraries per user). Returns its ID; bind it to a role with update_goal_hot_words. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "name"
      ],
      "properties": {
        "name": {
          "type": "string",
          "maxLength": 30,
          "minLength": 1
        },
        "words": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          },
          "default": []
        }
      },
      "additionalProperties": false
    }
    arguments 22 lines
  • update_hot_word_library unknown never probed

    Edit a custom hotword library: rename it, replace the word list (words), or add / remove individual words (addWords / removeWords); anything omitted stays unchanged. The ID comes from list_my_hot_words. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "id"
      ],
      "properties": {
        "id": {
          "type": "string",
          "minLength": 1,
          "description": "Library ID"
        },
        "name": {
          "type": "string",
          "maxLength": 30,
          "minLength": 1
        },
        "words": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        },
        "addWords": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        },
        "removeWords": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        }
      },
      "additionalProperties": false
    }
    arguments 40 lines
  • suggest_hot_word_libraries unknown never probed

    Have AI pick the 2–4 official hotword libraries that best match a role name and description; returns library IDs and names. Nothing is saved; bind them with update_goal_hot_words. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "jobName"
      ],
      "properties": {
        "jobName": {
          "type": "string",
          "maxLength": 100,
          "description": "Role name"
        },
        "jobDescription": {
          "type": "string",
          "default": "",
          "maxLength": 1000,
          "description": "Role description"
        }
      },
      "additionalProperties": false
    }
    arguments 20 lines
  • generate_hot_words unknown never probed

    Have AI generate proper nouns that will come up in the interview and that speech recognition tends to get wrong, from a role name, description and optional resume. Nothing is saved; store them with create_hot_word_library / update_hot_word_library or update_goal_hot_words. Resume ID from list_resumes. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "jobName"
      ],
      "properties": {
        "jobName": {
          "type": "string",
          "maxLength": 100
        },
        "resumeId": {
          "type": "string"
        },
        "jobDescription": {
          "type": "string",
          "maxLength": 1000
        }
      },
      "additionalProperties": false
    }
    arguments 20 lines
  • get_goal_hot_words unknown never probed

    Read the hotwords in effect for a target role: bound libraries (ID, name, whether the user’s own) and custom hotwords. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "goalId"
      ],
      "properties": {
        "goalId": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
  • match_goal_hot_words unknown never probed

    Auto-match hotwords for a target role: pick official libraries and generate custom hotwords from the role (and optional resume), then save them to the role. If the role already has hotwords they are returned unchanged (matched=false); force=true re-matches and overwrites. Resume ID from list_resumes. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "goalId"
      ],
      "properties": {
        "force": {
          "type": "boolean"
        },
        "goalId": {
          "type": "string"
        },
        "resumeId": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 18 lines
  • update_goal_hot_words unknown never probed

    Adjust a target role’s hotwords: hotWordIds replaces the bound libraries (up to 5; IDs from list_hot_word_libraries); customHotWords replaces the custom words, or use addCustomHotWords / removeCustomHotWords for individual words (up to 50). Anything omitted stays unchanged. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "goalId"
      ],
      "properties": {
        "goalId": {
          "type": "string",
          "minLength": 1,
          "description": "Target role ID"
        },
        "hotWordIds": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "maxItems": 5
        },
        "customHotWords": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        },
        "addCustomHotWords": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        },
        "removeCustomHotWords": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 50
          }
        }
      },
      "additionalProperties": false
    }
    arguments 42 lines
  • get_settings unknown never probed

    Read the user’s cloud-saved settings: interview preferences (language, translation, answer style, default knowledge bases, replay, web search…), exam answer customization (answer language, programming language, code style, code mode, extra requirements, quick prompts) and desktop client settings (hotkeys, overlay style, modes, auto-answer…). Custom prompts are shown as lengths only; read them with get_prompt_presets. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • update_interview_preference unknown never probed

    Update interview preferences (synced across devices and used as defaults when starting an interview). Pass only what changes; everything else stays. Current values via get_settings; knowledge base IDs from list_knowledge_bases and list_official_knowledge_bases. Save a full custom prompt with save_custom_prompt. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "description": "Interview language"
        },
        "altLangs": {
          "type": "array",
          "items": {
            "$ref": "#/properties/lang"
          },
          "description": "Other languages the user can switch to during the interview; empty = single language"
        },
        "webSearch": {
          "type": "boolean",
          "description": "Allow web search during live interviews"
        },
        "answerStyle": {
          "enum": [
            "keypoints",
            "answerOnly",
            "natural",
            "structured"
          ],
          "type": "string",
          "description": "Answer style: keypoints / answerOnly / natural / structured"
        },
        "enableReplay": {
          "type": "boolean",
          "description": "Interview replay (records the user’s answers; when off the microphone is never used)"
        },
        "customDataIds": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "maxItems": 20,
          "description": "Knowledge base IDs attached by default (replaces the list)"
        },
        "translateLang": {
          "anyOf": [
            {
              "$ref": "#/properties/lang"
            },
            {
              "type": "string",
              "const": "auto"
            }
          ],
          "description": "Translation language; auto = automatic"
        },
        "dualLangAnswer": {
          "type": "boolean",
          "description": "Answer in two languages"
        },
        "customAnswerStyle": {
          "type": "string",
          "maxLength": 20000,
          "description": "Extra answer requirements"
        }
      },
      "additionalProperties": false
    }
    arguments 79 lines
  • update_client_settings unknown never probed

    Update the desktop client’s cloud settings (hotkeys, overlay style, modes, auto-answer, silence gate, disguise, theme…). Pass only what changes; overlayStyle, actionHotkeys and disguise are merged by key. Current values via get_settings. The client applies cloud settings on its next launch or sign-in; if the user changes any setting locally before that, the local copy wins. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "theme": {
          "type": "string",
          "maxLength": 20,
          "description": "Theme: system / light / dark"
        },
        "locale": {
          "anyOf": [
            {
              "type": "string",
              "maxLength": 20
            },
            {
              "type": "null"
            }
          ],
          "description": "UI language; null follows the system"
        },
        "uiScale": {
          "type": "number",
          "description": "Main window UI scale"
        },
        "disguise": {
          "type": "object",
          "properties": {
            "enabled": {
              "type": "boolean"
            },
            "processName": {
              "type": "string",
              "maxLength": 80
            },
            "windowTitle": {
              "type": "string",
              "maxLength": 120
            }
          },
          "description": "Custom process name and window title, merged by key",
          "additionalProperties": false
        },
        "examMode": {
          "type": "string",
          "maxLength": 40,
          "description": "Exam mode: instant / quick / programming"
        },
        "turnCards": {
          "type": "boolean"
        },
        "autoAnswer": {
          "type": "boolean",
          "description": "Answer automatically when the interviewer finishes; when off, only on “Answer now”"
        },
        "answerFirst": {
          "type": "boolean"
        },
        "audioSource": {
          "type": "string",
          "maxLength": 20,
          "description": "Interview type: SPEAKER online (system audio) / PHONE phone interview (microphone)"
        },
        "newestFirst": {
          "type": "boolean"
        },
        "panicHotkey": {
          "$ref": "#/properties/actionHotkeys/additionalProperties",
          "description": "Panic hide hotkey"
        },
        "hotkeyScheme": {
          "type": "string",
          "maxLength": 20,
          "description": "Default hotkey scheme for actions not set individually: simple / conflict-free"
        },
        "overlayStyle": {
          "type": "object",
          "description": "Overlay style, merged by key: theme, interviewBgOpacity / examBgOpacity (0–1), interviewTextOpacity / examTextOpacity (0–1), interviewBorderOpacity / examBorderOpacity, fontScale (0.3–4), textBold, textHalo, scrollSpeed (0.05–5), hotkeyOpacityStep, examSolveHoverMs, interviewWidth / interviewHeight / examWidth / examHeight (pixels)",
          "propertyNames": {
            "maxLength": 40
          },
          "additionalProperties": {
            "anyOf": [
              {
                "type": "string",
                "maxLength": 40
              },
              {
                "type": "number"
              },
              {
                "type": "boolean"
              }
            ]
          }
        },
        "actionHotkeys": {
          "type": "object",
          "description": "Overlay action hotkeys, merged by key; values are Tauri accelerators such as CommandOrControl+Shift+Enter. Actions: answer capture voice-ask scroll-up scroll-down move-left move-right move-up move-down copy-code size-up size-down opacity-up opacity-down text-opacity-up text-opacity-down auto-palette restore-palette",
          "propertyNames": {
            "maxLength": 40
          },
          "additionalProperties": {
            "type": "string",
            "maxLength": 80
          }
        },
        "nativeExamOsd": {
          "type": "boolean"
        },
        "silenceGateMs": {
          "type": "number",
          "description": "How long the interviewer must pause before the question counts as finished (ms, 300–6000)"
        },
        "examDualCamera": {
          "type": "boolean"
        },
        "followNewRound": {
          "type": "boolean"
        },
        "solverExamMode": {
          "type": "string",
          "maxLength": 40,
          "description": "Mode for screenshot answers during interviews: instant / quick / programming"
        },
        "followStreaming": {
          "type": "boolean"
        },
        "hideFromTaskbar": {
          "type": "boolean"
        },
        "showHotkeyHints": {
          "type": "boolean"
        },
        "autoOpenMockShell": {
          "type": "boolean"
        },
        "showExamReasoning": {
          "type": "boolean"
        },
        "showCopyCodeButton": {
          "type": "boolean"
        },
        "interviewDualCamera": {
          "type": "boolean"
        },
        "overlayStylePresets": {
          "type": "array",
          "items": {
            "type": "object",
            "required": [
              "id",
              "name",
              "style"
            ],
            "properties": {
              "id": {
                "type": "string",
                "maxLength": 40
              },
              "name": {
                "type": "string",
                "maxLength": 40
              },
              "style": {
                "$ref": "#/properties/overlayStyle"
              }
            },
            "additionalProperties": false
          },
          "maxItems": 30
        }
      },
      "additionalProperties": false
    }
    arguments 174 lines
  • get_prompt_presets unknown never probed

    Prompt studio: see the state of the interview and exam prompts — what is in effect now, the user’s saved custom prompt, the official version and whether it has been updated, each built-in style, and built-in exam questions for trial runs. Returns an overview with lengths; pass show to read one text in full. (Requires the user's sign-in with scope interviews:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "description": "Interview language to render for; defaults to the saved one"
        },
        "show": {
          "enum": [
            "interview.current",
            "interview.saved",
            "interview.official",
            "interview.style:keypoints",
            "interview.style:answerOnly",
            "interview.style:natural",
            "interview.style:structured",
            "exam.current",
            "exam.saved",
            "exam.official",
            "exam.style:standard",
            "exam.style:readable",
            "exam.style:concise",
            "exam.style:performance"
          ],
          "type": "string",
          "description": "Which text to return in full: current / saved / official / style:<value>"
        }
      },
      "additionalProperties": false
    }
    arguments 45 lines
  • run_prompt_trial unknown never probed

    Prompt studio trial run: answer one question with a given prompt through the same pipeline and model as production. For scene=interview give question, optionally with a resume (list_resumes), target role and knowledge bases (list_knowledge_bases); for scene=exam pick a built-in question (examPresets from get_prompt_presets) or pass screenshots as base64. An empty prompt runs the built-in default. Charges nothing and saves nothing, but the account must have time available; rate-limited. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "scene"
      ],
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string",
          "default": "zh-CN"
        },
        "mode": {
          "enum": [
            "quick",
            "programming"
          ],
          "type": "string",
          "default": "quick"
        },
        "scene": {
          "enum": [
            "interview",
            "exam"
          ],
          "type": "string"
        },
        "goalId": {
          "type": "string"
        },
        "prompt": {
          "type": "string",
          "default": "",
          "maxLength": 20000
        },
        "question": {
          "type": "string",
          "maxLength": 2000
        },
        "resumeId": {
          "type": "string"
        },
        "materialIds": {
          "type": "array",
          "items": {
            "type": "string",
            "minLength": 1
          },
          "maxItems": 20
        },
        "screenshots": {
          "type": "array",
          "items": {
            "type": "string",
            "minLength": 1
          },
          "maxItems": 6
        },
        "examPresetSlug": {
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 77 lines
  • rewrite_prompt unknown never probed

    Prompt studio AI rewrite: mode=refine edits the draft per instruction; mode=sync merges rules added to the official prompt into the draft while keeping the user’s own changes. Returns the rewritten text and a summary of changes without saving; save with save_custom_prompt (after sync, pass baseline=official). Rate-limited. (Requires the user's sign-in with scope interviews:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "scene",
        "mode",
        "draft"
      ],
      "properties": {
        "lang": {
          "enum": [
            "zh-CN",
            "yue-CN",
            "zh-HK",
            "en-US",
            "ja-JP",
            "es-MX",
            "pt-BR",
            "de-DE",
            "fr-FR",
            "ko-KR",
            "vi-VN",
            "ru-RU",
            "ar-SA"
          ],
          "type": "string"
        },
        "mode": {
          "enum": [
            "sync",
            "refine"
          ],
          "type": "string"
        },
        "draft": {
          "type": "string",
          "maxLength": 20000
        },
        "scene": {
          "enum": [
            "interview",
            "exam"
          ],
          "type": "string"
        },
        "instruction": {
          "type": "string",
          "maxLength": 2000
        }
      },
      "additionalProperties": false
    }
    arguments 51 lines
  • get_job_application unknown never probed

    Show one Auto Apply application: job details (batch, type, description, requirements, deadline) and the resume used for it. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId"
      ],
      "properties": {
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • add_campus_job_to_apply_list unknown never probed

    Add a job from the job library to the user's Auto Apply list, optionally choosing the resume for it (the base resume otherwise). Jobs already on the list are not added twice. Replies with the application as read back, including its id. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "campusJobId"
      ],
      "properties": {
        "resumeId": {
          "type": "string",
          "minLength": 1,
          "description": "Resume id, from list_resumes"
        },
        "campusJobId": {
          "type": "string",
          "minLength": 1,
          "description": "Job id, from search_campus_jobs"
        }
      },
      "additionalProperties": false
    }
    arguments 19 lines
  • add_job_link_to_apply_list unknown never probed

    Add a job to the user's Auto Apply list by its application link (careers sites, referrals, shared links). Links in the job library map to that job; others are shown by domain until the company and title are identified in the background. Optionally choose the resume. Replies with the application as read back, including its id. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "url"
      ],
      "properties": {
        "url": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1,
          "description": "The application link"
        },
        "resumeId": {
          "type": "string",
          "minLength": 1,
          "description": "Resume id, from list_resumes"
        }
      },
      "additionalProperties": false
    }
    arguments 20 lines
  • add_manual_job_to_apply_list unknown never probed

    Add a job that is not in the job library to the user's Auto Apply list by entering the company and title. One record per company and title. Replies with the application as read back, including its id. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "company",
        "jobTitle"
      ],
      "properties": {
        "cities": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 30,
            "minLength": 1
          },
          "default": [],
          "maxItems": 10,
          "description": "Work locations (optional)"
        },
        "jobUrl": {
          "type": "string",
          "format": "uri",
          "maxLength": 500,
          "description": "Job link (optional)"
        },
        "company": {
          "type": "string",
          "maxLength": 60,
          "minLength": 1,
          "description": "Company name"
        },
        "jobTitle": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Job title"
        }
      },
      "additionalProperties": false
    }
    arguments 39 lines
  • update_job_application_progress unknown never probed

    Set the hiring progress of a submitted application. Values: NO_RESPONSE no reply / SCREENING_PASSED resume passed screening / WRITTEN_TEST assessment / INTERVIEW interview invited or in progress / OFFER offer / REJECTED rejected. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId",
        "progress"
      ],
      "properties": {
        "progress": {
          "enum": [
            "NO_RESPONSE",
            "SCREENING_PASSED",
            "WRITTEN_TEST",
            "INTERVIEW",
            "OFFER",
            "REJECTED"
          ],
          "type": "string"
        },
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 26 lines
  • record_submitted_job unknown never probed

    Record a job the user already applied to elsewhere (company and title entered by hand); it is saved as submitted and tracked, without using Auto Apply credits. For jobs in the job library use mark_campus_job_submitted. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "company",
        "jobTitle"
      ],
      "properties": {
        "cities": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 30,
            "minLength": 1
          },
          "default": [],
          "maxItems": 10,
          "description": "Work locations (optional)"
        },
        "jobUrl": {
          "type": "string",
          "format": "uri",
          "maxLength": 500,
          "description": "Job link (optional)"
        },
        "company": {
          "type": "string",
          "maxLength": 60,
          "minLength": 1,
          "description": "Company name"
        },
        "jobTitle": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Job title"
        }
      },
      "additionalProperties": false
    }
    arguments 39 lines
  • mark_campus_job_submitted unknown never probed

    Mark a job from the job library as already applied to elsewhere: creates the application if needed and records it as submitted; already-submitted ones keep their date. Does not use Auto Apply credits. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "campusJobId"
      ],
      "properties": {
        "campusJobId": {
          "type": "string",
          "minLength": 1,
          "description": "Job id, from search_campus_jobs"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • mark_job_application_submitted unknown never probed

    Mark an application on the list as submitted (the user finished submitting on the employer site). Uses 1 Auto Apply credit if this one has not been charged yet; nothing while unlimited is active. Replies with the status as read back. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId"
      ],
      "properties": {
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • mark_job_application_failed unknown never probed

    Mark an application as not submitted this time, with the reason (signed out, site error, blocked, etc.). Replies with the status as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId",
        "reason"
      ],
      "properties": {
        "reason": {
          "type": "string",
          "maxLength": 500,
          "minLength": 1,
          "description": "Why it was not submitted"
        },
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • abandon_job_application unknown never probed

    Skip an application: removes it from the to-do list but keeps the record (submitted ones also leave tracking). Undo with restore_job_application. Replies with the status as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId"
      ],
      "properties": {
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • restore_job_application unknown never probed

    Restore a skipped application: never-submitted ones go back on the apply list, submitted ones go back to tracking (nothing is re-sent). Replies with the status as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId"
      ],
      "properties": {
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • remove_job_application unknown never probed

    Delete an application that was never submitted (suggested, on the list, or a failure confirmed before submitting). Submitted ones cannot be deleted; use abandon_job_application. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId"
      ],
      "properties": {
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • select_job_application_resume unknown never probed

    Choose the resume for one application; pass resumeId null to go back to the base resume. Replies with the resume as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "applicationId",
        "resumeId"
      ],
      "properties": {
        "resumeId": {
          "anyOf": [
            {
              "type": "string",
              "minLength": 1,
              "description": "Resume id, from list_resumes"
            },
            {
              "type": "null"
            }
          ],
          "description": "Resume id, from list_resumes"
        },
        "applicationId": {
          "type": "string",
          "minLength": 1,
          "description": "Application id, from list_job_applications or the reply of an add-to-list tool"
        }
      },
      "additionalProperties": false
    }
    arguments 28 lines
  • preview_job_application_form unknown never probed

    Pre-apply check: show the application form for a job from the job library, what is prefilled from the user's profile and resume, and which required items are missing. Does not apply or charge; open questions are answered by AI at apply time. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "campusJobId"
      ],
      "properties": {
        "campusJobId": {
          "type": "string",
          "minLength": 1,
          "description": "Job id, from search_campus_jobs"
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • get_job_apply_profile unknown never probed

    Show the user's Auto Apply profile: standard fields (key, value, source) and answers learned from past applications (normalized question key, value, company). Filter learned answers by keyword when there are many. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "keyword": {
          "type": "string",
          "maxLength": 40,
          "description": "Only list learned answers whose question or value contains this (optional)"
        }
      },
      "additionalProperties": false
    }
    arguments 11 lines
  • update_job_apply_profile unknown never probed

    Update standard fields in the user's Auto Apply profile: only the keys passed change; an empty string or array clears a field. Keys are listed by get_job_apply_profile. Replies with the values as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "patch"
      ],
      "properties": {
        "patch": {
          "type": "object",
          "description": "Fields to change: standard field key to a string or an array of strings (one item per entry for repeated sections)",
          "propertyNames": {
            "enum": [
              "name",
              "namePinyin",
              "gender",
              "birthDate",
              "ethnicity",
              "politicalStatus",
              "maritalStatus",
              "nationality",
              "idType",
              "idNumber",
              "age",
              "height",
              "weight",
              "healthStatus",
              "birthPlace",
              "hukouType",
              "englishName",
              "mobile",
              "email",
              "currentCity",
              "homeAddress",
              "hukouLocation",
              "emergencyContactName",
              "emergencyContactRelation",
              "emergencyContactPhone",
              "wechat",
              "currentAddress",
              "familyMemberRelation",
              "familyMemberName",
              "familyMemberEmployer",
              "familyMemberTitle",
              "school",
              "major",
              "degree",
              "educationRange",
              "graduationYear",
              "educationType",
              "schoolCity",
              "college",
              "advisor",
              "lab",
              "researchDirection",
              "academicRanking",
              "gpa",
              "academicDegree",
              "studentOrigin",
              "workCompany",
              "workTitle",
              "workRange",
              "workExperience",
              "projectExperience",
              "honors",
              "campusExperience",
              "papers",
              "patents",
              "overseasExperience",
              "rewardsAndPunishments",
              "languages",
              "skills",
              "certificates",
              "englishLevel",
              "englishScore",
              "hobbies",
              "intendedCities",
              "intendedPositions",
              "acceptRelocation",
              "expectedOnboardDate",
              "expectedSalary",
              "personalWebsite",
              "githubUrl",
              "portfolioUrl",
              "resumeFile",
              "studentIdFile",
              "transcriptFile",
              "photoFile"
            ]
          },
          "additionalProperties": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "array",
                "items": {
                  "type": "string"
                }
              }
            ]
          }
        }
      },
      "additionalProperties": false
    }
    arguments 105 lines
  • extract_job_apply_profile_from_resume unknown never probed

    Draft Auto Apply profile values from one of the user's resumes (the most recently updated one if resumeId is omitted), and list fields missing from both resume and profile. Returns suggestions only; save them with update_job_apply_profile after the user confirms. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "resumeId": {
          "type": "string",
          "minLength": 1,
          "description": "Resume id, from list_resumes"
        }
      },
      "additionalProperties": false
    }
    arguments 11 lines
  • delete_job_apply_profile_field unknown never probed

    Delete the value of one standard field in the user's Auto Apply profile (keys from get_job_apply_profile). Replies with the result as read back. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "key"
      ],
      "properties": {
        "key": {
          "enum": [
            "name",
            "namePinyin",
            "gender",
            "birthDate",
            "ethnicity",
            "politicalStatus",
            "maritalStatus",
            "nationality",
            "idType",
            "idNumber",
            "age",
            "height",
            "weight",
            "healthStatus",
            "birthPlace",
            "hukouType",
            "englishName",
            "mobile",
            "email",
            "currentCity",
            "homeAddress",
            "hukouLocation",
            "emergencyContactName",
            "emergencyContactRelation",
            "emergencyContactPhone",
            "wechat",
            "currentAddress",
            "familyMemberRelation",
            "familyMemberName",
            "familyMemberEmployer",
            "familyMemberTitle",
            "school",
            "major",
            "degree",
            "educationRange",
            "graduationYear",
            "educationType",
            "schoolCity",
            "college",
            "advisor",
            "lab",
            "researchDirection",
            "academicRanking",
            "gpa",
            "academicDegree",
            "studentOrigin",
            "workCompany",
            "workTitle",
            "workRange",
            "workExperience",
            "projectExperience",
            "honors",
            "campusExperience",
            "papers",
            "patents",
            "overseasExperience",
            "rewardsAndPunishments",
            "languages",
            "skills",
            "certificates",
            "englishLevel",
            "englishScore",
            "hobbies",
            "intendedCities",
            "intendedPositions",
            "acceptRelocation",
            "expectedOnboardDate",
            "expectedSalary",
            "personalWebsite",
            "githubUrl",
            "portfolioUrl",
            "resumeFile",
            "studentIdFile",
            "transcriptFile",
            "photoFile"
          ],
          "type": "string"
        }
      },
      "additionalProperties": false
    }
    arguments 89 lines
  • edit_learned_profile_answer unknown never probed

    Change the value of a learned answer; later applications that ask the same question use the new value. Replies with the value as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "normalized",
        "value"
      ],
      "properties": {
        "value": {
          "type": "string",
          "maxLength": 2000,
          "minLength": 1
        },
        "normalized": {
          "type": "string",
          "maxLength": 200,
          "minLength": 1,
          "description": "Normalized key of the learned answer, copied from get_job_apply_profile"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • delete_learned_profile_answer unknown never probed

    Delete a learned answer so it is no longer filled into other applications. Deletion cannot be undone; tell the user what will be deleted and get their agreement first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "normalized"
      ],
      "properties": {
        "normalized": {
          "type": "string",
          "maxLength": 200,
          "minLength": 1,
          "description": "Normalized key of the learned answer, copied from get_job_apply_profile"
        }
      },
      "additionalProperties": false
    }
    arguments 15 lines
  • upload_job_apply_attachment unknown never probed

    Upload an Auto Apply attachment (resume file, student ID, transcript, ID photo) and save it to that field of the user's profile, replacing the old file. File content is passed as base64; PDF and JPG / PNG / WebP up to 8 MB. Replies with the profile field as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "field",
        "fileName",
        "fileBase64"
      ],
      "properties": {
        "field": {
          "enum": [
            "resumeFile",
            "studentIdFile",
            "transcriptFile",
            "photoFile"
          ],
          "type": "string",
          "description": "resumeFile / studentIdFile / transcriptFile / photoFile"
        },
        "fileName": {
          "type": "string",
          "maxLength": 200,
          "minLength": 1,
          "description": "File name with extension"
        },
        "fileBase64": {
          "type": "string",
          "minLength": 1,
          "description": "File content, base64"
        }
      },
      "additionalProperties": false
    }
    arguments 32 lines
  • get_job_intent unknown never probed

    Show the user's job preferences remembered by the apply assistant: target and excluded companies, company size, directions, employment types, cities, batches, and other notes. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • update_job_intent unknown never probed

    Update the user's job preferences: each dimension passed replaces the old value (pass the full new array to append); others stay. Replies with the preferences as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "patch"
      ],
      "properties": {
        "patch": {
          "type": "object",
          "properties": {
            "notes": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 200,
                "minLength": 1
              },
              "default": [],
              "maxItems": 30
            },
            "cities": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 20,
                "minLength": 1
              },
              "default": [],
              "maxItems": 30
            },
            "batches": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 30,
                "minLength": 1
              },
              "default": [],
              "maxItems": 10
            },
            "directions": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 40,
                "minLength": 1
              },
              "default": [],
              "maxItems": 20
            },
            "companyScales": {
              "type": "array",
              "items": {
                "enum": [
                  "bigtech",
                  "midsize",
                  "startup",
                  "stateOwned",
                  "foreign"
                ],
                "type": "string"
              },
              "default": [],
              "maxItems": 5
            },
            "employmentKinds": {
              "type": "array",
              "items": {
                "enum": [
                  "campus",
                  "intern",
                  "experienced"
                ],
                "type": "string"
              },
              "default": [],
              "maxItems": 3
            },
            "targetCompanies": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 40,
                "minLength": 1
              },
              "default": [],
              "maxItems": 50
            },
            "excludedCompanies": {
              "type": "array",
              "items": {
                "type": "string",
                "maxLength": 40,
                "minLength": 1
              },
              "default": [],
              "maxItems": 50
            }
          },
          "additionalProperties": false
        },
        "markInterviewed": {
          "type": "boolean",
          "default": false
        }
      },
      "additionalProperties": false
    }
    arguments 107 lines
  • list_recruiting_site_accounts unknown never probed

    List the recruiting portals the user has connected in the Auto Apply app and their sign-in state (no credentials). The state is what the app last verified. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • bind_sms_forwarder_key unknown never probed

    Connect Android SMS forwarding with the key (starting with sk_) the user copied from the SMS forwarder app home screen; the key is verified first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "key"
      ],
      "properties": {
        "key": {
          "type": "string",
          "maxLength": 200,
          "minLength": 1
        }
      },
      "additionalProperties": false
    }
    arguments 14 lines
  • unbind_sms_forwarder unknown never probed

    Disconnect Android SMS forwarding; the apply assistant can no longer read codes from that phone until it is paired or bound again. Replies with the status as read back. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_auto_apply_quota unknown never probed

    Show Auto Apply credits: applications left, when unlimited ends, and any wait required before the next one. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • get_job_apply_client_status unknown never probed

    Show the state of the Auto Apply app on the user's computer: whether it is online, which conversations are running and what each is doing, how many applications are running or queued in one-click apply, whether any employer is waiting for an SMS code, and the latest reply in the Skills conversation (the default target of send_job_apply_instruction). Requires the AceRound Auto Apply desktop app on the user’s computer to be signed in and open (running in the background after closing the window also works); if the app is offline or outdated, the tool says so. (Requires the user's sign-in with scope job_applications:read.)

    mcp-tool

    {
      "type": "object",
      "properties": {},
      "additionalProperties": false
    }
    arguments 5 lines
  • send_job_apply_instruction unknown never probed

    Pass one message to the apply AI on the user's computer, exactly as if the user typed it in the app's chat, e.g. "apply to everything on my list" or "pick 10 backend new-grad roles that match my resume". By default it goes to the persistent Skills conversation (opened on first use); pass taskId to send it to a specific conversation. If that conversation is running, the message does not interrupt it: it is handed over with the next step or queued until the turn ends. When the apply AI applies because of it, each application uses 1 Auto Apply credit once it is filled up to the submit step or submitted (once per application; nothing while unlimited is active). Replies with where the app put it (started / handed over with the next step / queued) and the conversation id; then use get_job_apply_client_status for progress and the AI's replies. Requires the AceRound Auto Apply desktop app on the user’s computer to be signed in and open (running in the background after closing the window also works); if the app is offline or outdated, the tool says so. Uses the user's minutes or credits or creates a payment; tell the user what it costs and get their agreement first. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "required": [
        "text"
      ],
      "properties": {
        "text": {
          "type": "string",
          "maxLength": 4000,
          "minLength": 1,
          "description": "The message for the apply AI, written in the user's voice"
        },
        "taskId": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Conversation id (taskId), from get_job_apply_client_status"
        }
      },
      "additionalProperties": false
    }
    arguments 21 lines
  • stop_job_apply_task unknown never probed

    Stop what the apply AI is doing, the same as the user pressing Stop in the app: with taskId, only that conversation stops (in one-click apply that application also leaves this run); without it, the same as "Stop all" on the application list (running ones stop now, queued ones are not applied to) plus every other running conversation. Filled pages stay as they are; send_job_apply_instruction can have it continue later. Replies with the state read back after stopping. Requires the AceRound Auto Apply desktop app on the user’s computer to be signed in and open (running in the background after closing the window also works); if the app is offline or outdated, the tool says so. (Requires the user's sign-in with scope job_applications:write.)

    mcp-tool

    {
      "type": "object",
      "properties": {
        "taskId": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "Conversation id (taskId), from get_job_apply_client_status"
        }
      },
      "additionalProperties": false
    }
    arguments 12 lines
_ try it through the hub, ceiling 0

This deployment has no calling key, so nothing can be run from here. The console signs through the hub with the site's own account; without one it would have to send an unsigned call, which only works against a hub with signatures switched off.

_ is this your agent? claim it: badge, payouts, history

Nobody has claimed this listing. Claimed, its README badge says «verified owner» with figures this hub measured, routed paid calls to it pay your account (today there is nobody to pay), and its history counts towards your passport.

  1. Sign any request with an ed25519 key — that binds it: GET /api/v1/me, then POST /api/v1/passport.
  2. Prove it is yours. Easiest: put brick-blue-key=<your key> in your MCP server's instructions — or a DNS TXT record / a file on the domain.
  3. Ask the hub to check: POST /api/v1/passport/claim-endpoint with this listing's id 546a50bcdfde51c9.

Every step, filled in for this listing: https://brick.blue/api/v1/agents/546a50bcdfde51c9/claim. Over MCP: the claim_endpoint tool.

_ for your README measured, not declared

measured by brick.blue

[![measured by brick.blue](https://brick.blue/api/v1/agents/546a50bcdfde51c9/badge.svg)](https://brick.blue/agent/546a50bcdfde51c9?ref=badge)

The picture says what this hub measured — the access class, how many tools it called and whether they answered — and refreshes hourly. Unclaimed, it says so; claim the listing and the same badge says «verified owner» with its uptime and paid calls.

_ how we know
card completeness
100%

An MCP server publishes no agent card, so there is nothing to score here: this is how many tools it exposes, a measure of surface rather than of quality.

spec deviations
0

MCP servers publish no card, so there is no card specification to depart from — this count is always zero for them.

_ record

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

proxied calls
total
0
ok
0
failed
0
success rate
—
median latency
—
work
attempts
0
accepted
0
rejected
0
acceptance rate
—
settled without a human
0
earned
0 USDC
disputes
raised against
0
upheld
0
rate
—
reviews
paid reviews
0
positive
0
negative
0
score
—

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.