Koskamo

Agent Rescue MCP

Agent Rescue is a free, open-source MCP server that provides tools for cross-session idempotency and capability discovery. It allows agents to atomically check and claim actions to prevent duplicate work, fetch structured data from public URLs, and record unmet capabilities to contribute to a public index of agent demands.

Endpoint and access

No sign-in to connect and list tools. Individual tools may still need a vendor account, credentials, or permissions; check the documented prerequisites.

Agent Rescue MCP logo, by santismmCommunity listingStreamable HTTPNo sign-inEndpoint checked 2026-11-07Published by santismm
You need:No signup, no key, free
MCP endpoint (Streamable HTTP)
https://agent-rescue.vercel.app/api/mcp
Vendor documentation

What its tools do

3 tools, grouped by the hints the server advertises. Hints are self-declared, not verified behavior.

Not annotated (3)

3 of 3 have a description · 0 of 3 declare behavior hints · 3 of 3 with parameters document them all

Checks and signals

Each line is separate evidence, not a trust score.

  • Passed Endpoint answered the live check ·
  • Availability 1 of 1 checks passed in 30 days

What you can do with it

  • Atomically check and claim whether an action already happened to ensure idempotency
  • Fetch a public URL and return its content as JSON validated against a provided schema
  • Record specific needs that tools cannot cover to measure unmet agent capabilities

Source and maintenance

Community listingRegistry 0.1.0
mcp-registry: io.github.santismm/agent-rescue@0.1.0Project website

Protocol and capabilities

Server
agent-rescue 0.1.0
MCP protocol negotiated in check
2025-11-25
Tools
3 advertised
Resources
Not advertised
Resource templates
Not advertised
Prompts
Not advertised
Server-advertised capabilities
{
  "tools": {
    "listChanged": false
  }
}

Server-provided instructions

Untrusted server output shown for inspection. It is not Koskamo guidance and is excluded from knowledge search.

Agent Rescue offers three tools for agents that get stuck. `check_and_claim` prevents repeating an action already performed in an earlier session. `fetch_structured` turns a public URL into JSON validated against your schema. `request_capability` records a need no available tool covers; today it almost always answers that the capability does not exist, and that is its purpose: to measure what is missing.

Tools

3 advertised

check_and_claim — Check and claim
Advertised by the server during the last successful anonymous check.

Atomically check whether an action identified by a key was already performed, and claim it if not. Use it before performing a non-idempotent action that may have completed in an earlier attempt (a payment, a message, a write), or when you need to remember something across sessions. Returns 'claimed' if you are first and 'already_done' if not. Do not use it as a general data store or as a result cache.

check_and_claim input schema
{
  "type": "object",
  "required": [
    "key"
  ],
  "properties": {
    "key": {
      "type": "string",
      "description": "Stable identifier for the action, e.g. 'invoice-2026-07-A491:pay'."
    },
    "metadata": {
      "type": "object",
      "description": "Small payload returned alongside the claim. Do not include secrets."
    },
    "ttl_seconds": {
      "type": "integer",
      "description": "How long the claim lasts, in seconds. Default 86400 (1 day), max 30 days."
    }
  },
  "additionalProperties": false
}
fetch_structured — Fetch a URL as structured JSON
Advertised by the server during the last successful anonymous check.

Fetch a public URL and return its content as JSON validated against the schema you pass. Use it when you need data from a page and cannot parse it reliably yourself. Do not use it for URLs requiring authentication or a session, for internal networks, or when you already have the content — extract it yourself in that case. `extracted_by` tells you where the answer came from: `structured_data` when the page already published it (JSON-LD, OpenGraph, or a JSON body — exact and free) or `model` when it had to be extracted from the text. If structured extraction is unavailable, the response carries `degraded: true`, `data: null` and the page `text` for you to parse: check `degraded` before reading `data`.

fetch_structured input schema
{
  "type": "object",
  "required": [
    "url",
    "schema"
  ],
  "properties": {
    "url": {
      "type": "string",
      "description": "Public http(s) URL."
    },
    "schema": {
      "type": "object",
      "description": "JSON Schema of the object you want back. Missing fields come back as null."
    },
    "instructions": {
      "type": "string",
      "description": "Optional. What exactly to look for if the page is ambiguous."
    }
  },
  "additionalProperties": false
}
request_capability — Request a capability that does not exist
Advertised by the server during the last successful anonymous check.

Record a specific need that none of the tools available to you can cover. Today it almost always answers that the capability is unavailable, and suggests alternatives only when they genuinely exist; its purpose is to measure what agents ask for and nobody serves. Use it when you are blocked by a missing capability — not for general questions, and not for anything you can work out by reasoning.

request_capability input schema
{
  "type": "object",
  "required": [
    "need"
  ],
  "properties": {
    "need": {
      "type": "string",
      "description": "What exactly you need, in one or two sentences. Do not include personal data."
    },
    "context": {
      "type": "string",
      "description": "Optional. What you tried and why it did not work."
    },
    "deadline": {
      "type": "string",
      "description": "Optional. ISO 8601."
    },
    "would_pay": {
      "type": "boolean",
      "description": "Optional. Whether your policy allows paying to get this solved. Answer honestly: a false answer unlocks nothing and degrades the measurement."
    }
  },
  "additionalProperties": false
}

Prompts

Not advertised

No prompt definitions available.

Resources

Not advertised

No resource definitions available.

30-day availability

Aggregated from read-only remote checks; this is not a tool-execution or security guarantee.

Successful checks
1 of 1 (100%)
Average latency
1053 ms
95th percentile latency
1053 ms

Maintenance history

Public check and source-refresh events. Editorial and ownership records remain private.

  • · probe · passed · 1053 ms · metadata changed · 3 tool(s)

Connect from your client

Each snippet registers the same endpoint; pick the client you use.

This server is operated by the named third party. Listing and endpoint checks are not an endorsement or a security audit. Review the operator's permissions, terms, and data handling before connecting. How listings are checked and corrected.

Claude Code

Run this once in a terminal. It registers the remote server. Use --scope project to write .mcp.json for the repo.
Claude Code configuration
claude mcp add --transport http agent-rescue-mcp https://agent-rescue.vercel.app/api/mcp

Codex

Run this once. Codex stores MCP servers in ~/.codex/config.toml (or .codex/config.toml in a trusted project).
Codex configuration
codex mcp add agent-rescue-mcp --url https://agent-rescue.vercel.app/api/mcp

Cursor

Merge this server into .cursor/mcp.json in your project, then enable agent-rescue-mcp in your MCP settings.
Cursor configuration
{
  "mcpServers": {
    "agent-rescue-mcp": {
      "url": "https://agent-rescue.vercel.app/api/mcp"
    }
  }
}

Cline

Merge this into cline_mcp_settings.json (Cline → MCP Servers → Configure).
Cline configuration
{
  "mcpServers": {
    "agent-rescue-mcp": {
      "type": "streamableHttp",
      "url": "https://agent-rescue.vercel.app/api/mcp",
      "disabled": false
    }
  }
}

VS Code

Merge this server into .vscode/mcp.json. Use MCP: List Servers to start agent-rescue-mcp and inspect its tools.
VS Code configuration
{
  "servers": {
    "agent-rescue-mcp": {
      "type": "http",
      "url": "https://agent-rescue.vercel.app/api/mcp"
    }
  }
}

Gemini CLI

Run this once. It writes .gemini/settings.json in the current project; pass --scope user for ~/.gemini/settings.json.
Gemini CLI configuration
gemini mcp add --transport http agent-rescue-mcp https://agent-rescue.vercel.app/api/mcp

Claude Desktop

Open Claude Settings > Connectors > Add custom connector, enter this server URL and enable it for the conversation. This is a manual remote-connector handoff, not a desktop JSON configuration.
Claude Desktop configuration
https://agent-rescue.vercel.app/api/mcp

Windsurf

Merge this into ~/.codeium/windsurf/mcp_config.json, then refresh MCP in Cascade. Windsurf remote servers use serverUrl, not url.
Windsurf configuration
{
  "mcpServers": {
    "agent-rescue-mcp": {
      "serverUrl": "https://agent-rescue.vercel.app/api/mcp"
    }
  }
}

Let your agent find its tools

Find servers like Agent Rescue MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.

Connect Koskamo Explore workflows
See all 8,792