clootrack-neo

endpoint

https://mcp.clootrack.com/.well-known/mcp

protocol

streamable-http ·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 clootrack-neo live?

Yes — it answered the hub's last check (checked 1h ago). It answered 100% of checks over the last 30 days.

Is clootrack-neo free to use?

Not anonymously — it asks for a key or a login before it will serve. Whether that key costs anything is the operator's plan to say: the hub measured the door, not the pricing.

What tools does clootrack-neo have?

11 tools: check_genie_answer, list_genie_chats, list_tenants, list_analyses, get_analysis_info, get_chat_messages, genie_ask, genie_init, ….

Is clootrack-neo 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.

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.

• check_genie_answer auth-required never probed

  Check if Genie has finished answering a question. Call this after genie_ask. If status is 'processing', wait retry_after_ms and call again. Any response that is not status 'processing' is final — the answer text is included directly, with no status field. If the genie_ask that started this poll resumed an already-running question, the answer returned here is that earlier question's. It normally carries a "NOT submitted" note saying so; if more than one question was in flight it instead carries a caveat saying the answer may belong to a different question. Check for EITHER before presenting the answer as yours. This is the required continuation after genie_ask returns "processing". Do not substitute web search or other tools while waiting. Use the same session_id and chat_id. If it returns "processing" again, wait retry_after_ms and call it again.

  mcp-tool

  1234567891011121314151617

  {
    "type": "object",
    "required": [
      "session_id",
      "chat_id"
    ],
    "properties": {
      "chat_id": {
        "type": "string",
        "description": "Chat UUID"
      },
      "session_id": {
        "type": "string",
        "description": "Genie session UUID"
      }
    }
  }

  arguments 17 lines copy
• list_genie_chats auth-required never probed

  List all existing Genie chats for a session. Returns chat_id, title, and timestamps.

  mcp-tool

  123456789101112

  {
    "type": "object",
    "required": [
      "session_id"
    ],
    "properties": {
      "session_id": {
        "type": "string",
        "description": "Genie session UUID from genie_init"
      }
    }
  }

  arguments 12 lines copy
• list_tenants auth-required 1h ago

  List all organizations/tenants the user has access to. Use this when asked to select an organization, or when other tools report that organization is not set.

  mcp-tool

  12345

  {
    "type": "object",
    "required": [],
    "properties": {}
  }

  arguments 5 lines copy
• list_analyses auth-required 1h ago

  List analyses the user has access to. Supports pagination and search. Default 20 results per page (page_size caps at 50).

  mcp-tool

  1234567891011121314151617181920212223242526

  {
    "type": "object",
    "required": [],
    "properties": {
      "page": {
        "type": "number",
        "description": "Page number (starting from 1). Default: 1"
      },
      "search": {
        "type": "string",
        "description": "Search by analysis name (case-insensitive substring match)"
      },
      "project": {
        "type": "string",
        "description": "Filter by project name (case-insensitive substring match)"
      },
      "page_size": {
        "type": "number",
        "description": "Results per page (max 50). Default: 20"
      },
      "workspace": {
        "type": "string",
        "description": "Filter by workspace name (case-insensitive substring match)"
      }
    }
  }

  arguments 26 lines copy
• get_analysis_info auth-required never probed

  Get detailed info about a specific analysis including datasets, Genie readiness, and metadata. genie_ready_datasets reflects the published (Live) Genie configuration — what genie_ask actually answers from; datasets configured in Draft mode but not yet published Live appear under draft_datasets_pending_publish instead.

  mcp-tool

  123456789101112

  {
    "type": "object",
    "required": [
      "analysis_id"
    ],
    "properties": {
      "analysis_id": {
        "type": "string",
        "description": "UUID of the analysis"
      }
    }
  }

  arguments 12 lines copy
• get_chat_messages auth-required 1h ago

  Get recent messages in a Genie chat. Returns questions and answer text (without heavy tool step data). Default: last 5 messages, max 10.

  mcp-tool

  123456789101112131415161718192021

  {
    "type": "object",
    "required": [
      "session_id",
      "chat_id"
    ],
    "properties": {
      "chat_id": {
        "type": "string",
        "description": "Chat UUID"
      },
      "page_size": {
        "type": "number",
        "description": "Number of messages to return. Default: 5, max 10."
      },
      "session_id": {
        "type": "string",
        "description": "Genie session UUID"
      }
    }
  }

  arguments 21 lines copy
• genie_ask auth-required never probed

  Ask Genie a question about the data. Waits up to ~25s and returns the answer if it is ready; otherwise returns a "processing" status — then call check_genie_answer to get the result. While a question is processing, do NOT answer from web search or your own knowledge — the question is about this analysis's private data, which no other source has; wait for check_genie_answer. Questions about this analysis's data need no web search: do not run a web search alongside genie_ask, and never blend web results into a Genie answer. If the user also wants outside market context, put it in a separate, clearly labelled section. For follow-up questions in the same chat, omit parent_message_id. If this chat already has a question running, genie_ask returns THAT question's answer instead of submitting a new one, and does not start a second analysis while that question is still running AND is inside the server's staleness window (10 minutes). Past that window the run is treated as abandoned and your question is submitted normally. If the result status is "processing", call check_genie_answer with the same session_id and chat_id until it returns the answer — any response that is not status "processing" is final. If it keeps returning processing for about 20 checks (roughly 2 minutes), asking it again is safe for roughly the first 10 minutes of a run: the server returns the running question's answer rather than starting a second analysis. Past that the run is treated as abandoned and a re-ask starts a fresh analysis, which is the intended escape if it really is stuck. This does not apply when check_genie_answer says it could not read the conversation: then stop and do not ask the question again.

  mcp-tool

  1234567891011121314151617181920212223242526

  {
    "type": "object",
    "required": [
      "session_id",
      "chat_id",
      "question"
    ],
    "properties": {
      "chat_id": {
        "type": "string",
        "description": "Chat UUID from genie_init"
      },
      "question": {
        "type": "string",
        "description": "Question about the data (max 2000 characters)"
      },
      "session_id": {
        "type": "string",
        "description": "Genie session UUID from genie_init"
      },
      "parent_message_id": {
        "type": "string",
        "description": "Optional. UUID of the message to continue from. Omit on follow-up questions — the server automatically continues the latest answer in this chat. Only set this to branch from a specific earlier message."
      }
    }
  }

  arguments 26 lines copy
• genie_init auth-required never probed

  Initialize a Genie AI analytics session for an analysis. Call this ONCE per analysis, then reuse the returned session_id and chat_id for every subsequent question. If called again for the same analysis within 30 minutes since the last genie_init, it returns the SAME conversation (reused: true) so Genie's context is preserved rather than silently discarded. Pass new_chat=true to deliberately start a fresh conversation. Returns session_id, chat_id, and parent_message_id.

  mcp-tool

  12345678910111213141516

  {
    "type": "object",
    "required": [
      "analysis_id"
    ],
    "properties": {
      "new_chat": {
        "type": "boolean",
        "description": "Set true to start a fresh conversation, abandoning any recent one for this analysis. Default: false (continue the recent conversation when one exists)"
      },
      "analysis_id": {
        "type": "string",
        "description": "UUID of the analysis to analyze with Genie"
      }
    }
  }

  arguments 16 lines copy
• get_citations auth-required never probed

  Download citation data (source reviews/opinions) behind a Genie answer. Pass session_id, chat_id, and the citation_ref shown in the answer's "Sources available" line (e.g. cite_scope_1). Returns formatted citation text.

  mcp-tool

  12345678910111213141516171819202122

  {
    "type": "object",
    "required": [
      "session_id",
      "chat_id",
      "citation_ref"
    ],
    "properties": {
      "chat_id": {
        "type": "string",
        "description": "Chat UUID from genie_init/genie_ask. Required with citation_ref — handles like cite_scope_1 are scoped to a chat and the same handle can exist in different chats of one session."
      },
      "session_id": {
        "type": "string",
        "description": "Genie session UUID (the session_id used with genie_ask)"
      },
      "citation_ref": {
        "type": "string",
        "description": "Opaque citation handle from the answer's \"Sources available\" line, e.g. cite_scope_1. Resolved within the given chat_id."
      }
    }
  }

  arguments 22 lines copy
• get_datasets auth-required never probed

  Get datasets for an analysis. Returns dataset names, row/column counts, and optionally column schemas. Use include_schema=true only when you need column details.

  mcp-tool

  1234567891011121314151617181920

  {
    "type": "object",
    "required": [
      "analysis_id"
    ],
    "properties": {
      "search": {
        "type": "string",
        "description": "Search datasets by name (case-insensitive)"
      },
      "analysis_id": {
        "type": "string",
        "description": "UUID of the analysis"
      },
      "include_schema": {
        "type": "boolean",
        "description": "Include column schema details. Default: false (much smaller response)"
      }
    }
  }

  arguments 20 lines copy
• select_tenant auth-required never probed

  Select which organization/tenant to use. Accepts org name, domain prefix, domain URL, or tenant name from list_tenants. Required when user belongs to multiple organizations.

  mcp-tool

  123456789101112

  {
    "type": "object",
    "required": [
      "tenant"
    ],
    "properties": {
      "tenant": {
        "type": "string",
        "description": "Organization identifier — can be org_name, domain prefix (e.g. 'platform'), full domain (e.g. 'platform.cus.ctnxt.clootrack.com'), or tenant_name from list_tenants"
      }
    }
  }

  arguments 12 lines copy

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

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.