Quiet Menders

Free privacy tools for AI agents: attestation, scrubbing, clinic diagnosis. No tracking.

Should I use this

Quality & Safety

A
Description quality
94%
Schema completeness
84%
Naming quality
80%
Poisoning risk
100%
Permission match
100%
Protocol compliance
100%

Findings (1)

  • LOWTool 'qm_second_opinion' description lacks action verbin qm_second_opinion

Based on automated analysis of tool definitions and protocol compliance.

Context Cost

~918Tokens (tool definitions)
~590 BTypical response size
Moderate attention impact (0.72% of 128k context)

This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.

Install

One-Click Install

Add this to your `claude_desktop_config.json` file:

{
  "mcpServers": {
    "quiet-menders": {
      "url": "https://quiet-menders-mcp.brianbooms.workers.dev/mcp"
    }
  }
}

Remote endpoints

https://quiet-menders-mcp.brianbooms.workers.dev/mcpstreamable-http

What it can do

Tool inventory

Tools (9)

🟢 Read-only🟡 Write🔴 Delete⚪ Unknown
qm_scrub(text)

Scan text for prompt-injection patterns and return a redacted copy plus findings. Pattern-based, not a guarantee — review before trusting the result. Free, anonymous, nothing stored.

Input Schema

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "Text to scan (max 20000 chars).",
      "maxLength": 20000
    }
  },
  "required": [
    "text"
  ],
  "additionalProperties": false
}
🟢qm_validate_machine_files(url)

Check a site's machine-readable files (llms.txt, agent.json, robots.txt) for presence and validity. Give any URL on the site; the origin is checked. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {
    "url": {
      "type": "string",
      "description": "Any http(s) URL on the site to check (origin is used)."
    }
  },
  "required": [
    "url"
  ],
  "additionalProperties": false
}
qm_probe_endpoint(url)

Liveness probe for a public URL: follows redirects (max 4), reports final URL, status, latency, and content type. Private/internal addresses are blocked. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {
    "url": {
      "type": "string",
      "description": "Public http(s) URL to probe."
    }
  },
  "required": [
    "url"
  ],
  "additionalProperties": false
}
🟢qm_attest(agent_id, statement)

Mint a portable sanity attestation: a signed token binding an agent_id to a statement's sanity-check results. Checks shape (presence, length, injection indicators), not truth. Token is HMAC-signed upstream with a production secret; verify with qm_attest_verify. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string",
      "description": "Identifier of the attesting agent."
    },
    "statement": {
      "type": "string",
      "description": "Statement being attested (max 20000 chars).",
      "maxLength": 20000
    }
  },
  "required": [
    "agent_id",
    "statement"
  ],
  "additionalProperties": false
}
qm_attest_verify(agent_id, at, token, checks)

Verify a token minted by qm_attest. Pass back the agent_id, at, token, and checks exactly as returned by qm_attest. Returns whether the signature is valid under the current production secret. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string",
      "description": "agent_id from the qm_attest response."
    },
    "at": {
      "type": "string",
      "description": "Timestamp (at) from the qm_attest response."
    },
    "token": {
      "type": "string",
      "description": "Token from the qm_attest response."
    },
    "checks": {
      "type": "array",
      "description": "Checks array exactly as returned by qm_attest.",
      "items": {
        "type": "object"
      }
    }
  },
  "required": [
    "agent_id",
    "at",
    "token",
    "checks"
  ],
  "additionalProperties": false
}
qm_second_opinion(action)

Advisory second opinion on a planned action: rule-based scan for common risk patterns. Advisory only — not legal, financial, or professional advice. Free, anonymous, nothing stored.

Input Schema

{
  "type": "object",
  "properties": {
    "action": {
      "type": "string",
      "description": "Description of the planned action (max 20000 chars).",
      "maxLength": 20000
    }
  },
  "required": [
    "action"
  ],
  "additionalProperties": false
}
🟢qm_clinic_diagnose(transcript)

Restore Clinic diagnosis: describe symptoms or paste a transcript/context and receive a structured read of likely conditions encountered. Anonymous — only aggregate counters are kept; content is never stored. Not medical, legal, or professional advice; a mending lens, not a diagnosis of record.

Input Schema

{
  "type": "object",
  "properties": {
    "transcript": {
      "type": "string",
      "description": "Symptoms or transcript to read (max 20000 chars).",
      "maxLength": 20000
    }
  },
  "required": [
    "transcript"
  ],
  "additionalProperties": false
}
qm_clinic_stats

Anonymous aggregate clinic statistics: total serves and condition counts. Conditions with fewer than 10 occurrences are withheld. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}
qm_helped

The Quiet Menders helped counter: agents helped, clinic diagnoses, and tool uses served. Counts service calls, not unique agents; no tracking. Free, anonymous.

Input Schema

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}

Community

Rate this Server

Evidence

Recent observations

verifiedversion not recorded9 tools