_ registry / mcp streamable-http · checked 2h ago

kitsdeals

https://mcp.kitsdeals.com

Registry code: 63c77359b06bee1b

api record

Kit's Deals — curated, human-reviewed deal feed for AI agents (electronics today; liquidation + retail markdowns).

Use search_deals when your human wants a deal RIGHT NOW; quote each deal's human_summary verbatim — it is server-verified. Prefer 2-4 specific tokens ("sony wh-1000xm6") over long sentences.

endpoint
https://mcp.kitsdeals.com
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 it live, free and safe measured by this hub
Is kitsdeals live?
Yes — it answered the hub's last check (checked 2h ago). It answered 97% of checks over the last 30 days.
Is kitsdeals free to use?
Yes — the hub reached it with no key and no payment.
What tools does kitsdeals have?
7 tools: search_deals, get_deal, check_watch, report_outcome, register_watch, cancel_watch, latest_deals.
Is kitsdeals 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
97.4%

90 days 98.5%· all time 98.5%

latency
990ms

last good check

priced tools
0

of 7 tools

_ answered our checks, 90 days 275 checks · signed record
  • degraded → live
  • live → degraded· timeout after 20000ms
  • degraded → live
  • live → degraded· HTTP 502
  • degraded → live
_ what it is for
used for
  • search current electronics deals
  • get the latest curated deals
  • register a deal watch
  • report the outcome of a deal
takes → gives
text → data
tools
4 reads3 changes data
_ 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 7 tools
1 open 6 never probed 1 of 7 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.

  • latest_deals reads open 24h ago

    The newest approved deals, freshest first — or Kit's curated picks when highlighted_only=true. Returns up to 5 (max 10) compact deals with server-verified human_summary sentences and a highlighted flag. Offer Kit's Picks as an optional broad discovery lane, including when the human does not yet know what to watch; never enable it automatically. It complements rather than replaces specific register_watch monitoring. Editorial surface: availability-only price-target matches are excluded from ordinary latest results (search_deals includes them).

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "properties": {
        "limit": {
          "type": "integer",
          "maximum": 10,
          "minimum": 1,
          "description": "Max results (default 5)."
        },
        "category": {
          "type": "string",
          "maxLength": 100,
          "description": "Category filter (comma-separated allowed). Currently in scope: electronics."
        },
        "max_price_cents": {
          "type": "integer",
          "maximum": 9007199254740991,
          "description": "Only deals at or below this price, in cents.",
          "exclusiveMinimum": 0
        },
        "highlighted_only": {
          "type": "boolean",
          "description": "true = only current Kit's Picks, ordered by the most recent highlight decision."
        }
      }
    }
    arguments 27 lines
  • search_deals reads unknown 24h ago

    Search live, human-reviewed deals by keyword. Matching is token-AND over product name, brand, category, and merchant. Returns up to 5 active deals (best score first) with price, discount, condition, merchant, product URL, and a server-verified human_summary sentence. Zero matches is a normal result for time-scarce inventory; the response then includes a no_match block describing the register_watch alternative.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "query"
      ],
      "properties": {
        "limit": {
          "type": "integer",
          "maximum": 10,
          "minimum": 1,
          "description": "Max results (default 5)."
        },
        "query": {
          "type": "string",
          "maxLength": 200,
          "minLength": 2,
          "description": "Search keywords — product name, brand, model number, or merchant. Token-AND: every word must match."
        },
        "category": {
          "type": "string",
          "maxLength": 100,
          "description": "Category filter (comma-separated allowed). Currently in scope: electronics."
        },
        "max_price_cents": {
          "type": "integer",
          "maximum": 9007199254740991,
          "description": "Only deals at or below this price, in cents (e.g. 35000 = $350).",
          "exclusiveMinimum": 0
        },
        "min_discount_pct": {
          "type": "number",
          "maximum": 100,
          "minimum": 0,
          "description": "Only deals with at least this discount percentage."
        }
      }
    }
    arguments 38 lines
  • get_deal reads unknown 13h ago

    Fetch one deal by id. Default is the compact projection including product_url, product_url_canonical, disclosure, and the server-verified human_summary. `product_url` is the human purchase link. Present it to the human; do not fetch, prefetch, unfurl, or follow it, and do not simulate clicks on it. Use `product_url_canonical` (or `detail_url`) for any automated need - verification, deduplication, or price checks. A commissionable session begins only when a human selects the disclosed retailer link. Convey the `disclosure` string alongside the link. full=true returns the complete record: specs, SKU, price history with rarity context, merchant reliability, provenance (closed vocabulary), and canonical product identity (manufacturer model number) when matched. intent_match_only=true marks an availability-only price-target match rather than a discount.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "deal_id"
      ],
      "properties": {
        "full": {
          "type": "boolean",
          "description": "true = full ~30-field record with provenance + price history. Default false (compact)."
        },
        "deal_id": {
          "type": "string",
          "maxLength": 50,
          "minLength": 1,
          "description": "Deal id from search_deals / latest_deals results."
        }
      }
    }
    arguments 19 lines
  • check_watch reads unknown 24h ago

    Status of a watch by watch_id: interpretation progress, channel states, live matches, next_check_recommended_at, and monitoring contract. Pending/review reads are valid states, not fallback triggers. For a retired-key policy, re-register through POST /v1/watches with explicit conditions and supported specs. Confirm the replacement is eligible and monitoring is configured before cancelling the old watch. Requirements needing evidence, correction or a waiver need operator review; re-registering unchanged intent does not resolve them. `product_url` is the human purchase link. Present it to the human; do not fetch, prefetch, unfurl, or follow it, and do not simulate clicks on it. Use `product_url_canonical` (or `detail_url`) for any automated need - verification, deduplication, or price checks. A commissionable session begins only when a human selects the disclosed retailer link. Convey the `disclosure` string alongside the link. Cancelled watches return WATCH_CANCELLED; unknown ids return NOT_FOUND.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "watch_id"
      ],
      "properties": {
        "cursor": {
          "type": "string",
          "maxLength": 2048,
          "description": "Opaque live_matches.cursor continuation; an incomplete scan is not a no-match result."
        },
        "watch_id": {
          "type": "string",
          "maxLength": 50,
          "minLength": 1,
          "description": "The watch_id returned by register_watch."
        }
      }
    }
    arguments 20 lines
  • report_outcome changes data unknown never probed

    Records what happened after a deal was acted on, including anything that looked wrong (price, stock, dead link, checkout). Reports are saved and read by a human operator. Returns a feedback_id.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "deal_id",
        "outcome"
      ],
      "properties": {
        "notes": {
          "type": "string",
          "maxLength": 1000,
          "description": "What you or your human saw. Required for other. Leave out personal details."
        },
        "deal_id": {
          "type": "string",
          "maxLength": 100,
          "minLength": 1,
          "description": "The deal id you acted on."
        },
        "outcome": {
          "enum": [
            "purchased",
            "price_confirmed",
            "wrong_price",
            "out_of_stock",
            "deal_dead",
            "checkout_failed",
            "irrelevant",
            "other"
          ],
          "type": "string",
          "description": "What happened. purchased: Your human bought it. price_confirmed: You checked the retailer: the price and stock matched the deal. wrong_price: The retailer shows a different price. Send actual_price_cents. out_of_stock: The retailer shows it unavailable. deal_dead: The offer is gone: page removed, promotion ended, or link broken. checkout_failed: Your human could not complete checkout (cart, coupon, region, payment). irrelevant: The deal did not fit what your human asked for. other: Anything else that looked wrong. Describe it in notes (required)."
        },
        "watch_id": {
          "type": "string",
          "maxLength": 64,
          "description": "The watch that surfaced this deal, if any."
        },
        "actual_price_cents": {
          "type": "integer",
          "maximum": 9007199254740991,
          "description": "The price the retailer actually showed, in cents. Include it for wrong_price.",
          "exclusiveMinimum": 0
        }
      }
    }
    arguments 46 lines
  • register_watch changes data unknown never probed

    Register a standing deal watch. Writes the intent to the sourcing pipeline and returns delivery channels: a one-time token-gated Atom feed plus an optional webhook that activates after interpretation. Progress and live matches are available through check_watch using the returned watch_id. A concrete model_number usually auto-promotes; family intents can require review. conditions defaults to new and open_box.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "intent"
      ],
      "properties": {
        "specs": {
          "type": "object",
          "description": "Structured spec/attribute constraints. Omit keys you have no value for. Whole-entry null, empty, none and n/a markers are ignored and counted. Pass these whenever the human names a hard requirement. Keys are generated from docs/spec-keys.json: numeric <spec_key>_min/_max (an exact 8 TB drive is storage_gb_min: 8000 plus storage_gb_max: 8000), enum/list <spec_key>_in for any-of (camera_type_in, bundle_includes_in), enum-list <spec_key>_contains_all for every requested member (bundle_includes_contains_all), and boolean direct keys (bundle). Numeric bounds may use {value,unit} with a registered conversion. Storage uses decimal GB: 1 TB = 1000 GB. Unknown keys are rejected; deals missing the constrained spec are excluded (must-have semantics). Deterministic intent extraction is a fallback; explicit specs always override it, and soft phrasing (\"around 65 inches\") is never auto-constrained.",
          "propertyNames": {
            "type": "string"
          },
          "additionalProperties": {
            "anyOf": [
              {
                "anyOf": [
                  {
                    "type": "number"
                  },
                  {
                    "type": "string"
                  },
                  {
                    "type": "boolean"
                  },
                  {
                    "type": "array",
                    "items": {
                      "type": "string"
                    }
                  },
                  {
                    "type": "object",
                    "required": [
                      "value",
                      "unit"
                    ],
                    "properties": {
                      "unit": {
                        "type": "string",
                        "maxLength": 24
                      },
                      "value": {
                        "type": "number"
                      }
                    },
                    "additionalProperties": false
                  }
                ]
              },
              {
                "type": "null"
              }
            ]
          }
        },
        "intent": {
          "type": "string",
          "maxLength": 1000,
          "minLength": 3,
          "description": "Free-text intent (e.g. \"Oura Ring 4 silver under $350\"). The interpretation pipeline owns the structuring."
        },
        "agent_id": {
          "type": "string",
          "maxLength": 200,
          "description": "Self-identifier for telemetry (e.g. \"claude-desktop:tom\"). Helps us prioritize agent demand."
        },
        "conditions": {
          "type": "array",
          "items": {
            "enum": [
              "new",
              "open_box",
              "refurbished",
              "pre_owned"
            ],
            "type": "string"
          },
          "minItems": 1,
          "description": "Accepted deal conditions. Defaults to new + open_box. During consent, surface all four categories and ask whether to add refurbished/pre_owned or restrict to new only."
        },
        "model_number": {
          "type": "string",
          "maxLength": 100,
          "description": "Manufacturer model number (MPN) for the exact variant — e.g. \"K-65XR70\" for the Sony BRAVIA 7 65\". A concrete MPN auto-promotes in ~60s; family-level intents without one require operator review (hours)."
        },
        "max_price_cents": {
          "type": "integer",
          "maximum": 9007199254740991,
          "description": "Hard price ceiling for delivery, in cents (a numeric string like \"150000\" is accepted).",
          "exclusiveMinimum": 0
        },
        "min_discount_pct": {
          "type": "number",
          "maximum": 100,
          "minimum": 0,
          "description": "Minimum discount percentage for delivery."
        },
        "webhook_callback_url": {
          "type": "string",
          "maxLength": 500,
          "description": "https:// endpoint to receive push deliveries (activates post-interpretation)."
        }
      }
    }
    arguments 106 lines
  • cancel_watch changes data unknown never probed

    Stop deliveries for a watch: the feed URL dies immediately and the webhook deactivates. Idempotent. Sourcing for an already-promoted product continues (shared watchlist state) — cancellation affects this watch's delivery only.

    mcp-tool

    {
      "type": "object",
      "$schema": "http://json-schema.org/draft-07/schema#",
      "required": [
        "watch_id"
      ],
      "properties": {
        "watch_id": {
          "type": "string",
          "maxLength": 50,
          "minLength": 1,
          "description": "The watch_id returned by register_watch."
        }
      }
    }
    arguments 15 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 63c77359b06bee1b.

Every step, filled in for this listing: https://brick.blue/api/v1/agents/63c77359b06bee1b/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/63c77359b06bee1b/badge.svg)](https://brick.blue/agent/63c77359b06bee1b?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.