Skip to main content
Glama
ThoughtProof

thoughtproof-mcp

Official
by ThoughtProof
OverviewSchemaRelated ServersScoreDiscussions
JavaScript
Remote

thoughtproof-mcp

npm version CI License: MIT

thoughtproof-mcp — local stdio. Hero tool verify_decision (DQL spend / Sentinel irreversible exit). execute is true only on ALLOW.

MCP server for ThoughtProof — pre-execution decision verification for AI agents.

Hero tool: verify_decision. It routes inside the tool to DQL (spend / checkout) or Sentinel (irreversible exit) and returns a fail-closed execute flag. execute is true only on a native ALLOW.

This package is a local stdio MCP server (Node 18+) for Desktop / CLI hosts such as Cursor, Claude Desktop, Windsurf, and Cline. It is not a remote HTTP MCP server. It is not a Grok Web/Mobile custom connector.

Get keys at https://app.thoughtproof.ai/pricing.

Unpublished work is documented in UNRELEASED.md.

Quick Start

{
  "mcpServers": {
    "thoughtproof": {
      "command": "npx",
      "args": ["-y", "thoughtproof-mcp@0.3.2"],
      "env": {
        "DQL_API_KEY": "dqlk_your_key_here"
      }
    }
  }
}

Install with npx -y thoughtproof-mcp@0.3.2. Works with Claude Desktop, Cursor, Windsurf, Cline, and other local stdio MCP clients.

Related MCP server: agentshield-mcp

Tools

verify_before_action / verify_decision (hero)

verify_before_action is an alias of verify_decision (identical schema + handler). Soft fail-closed: host must honor execute=false.

Pre-execution gate for a proposed action. Routing is inside the tool — not an agent quiz.

Parameter

Type

Default

Description

mandate

string

(required)

User's stated goal / instruction

proposed_action

string

(required)

What the agent is about to do

reasoning

string

(required)

The agent's own plan / reasoning

context

string

(optional)

Extra evidence

mode

dql / sentinel / auto

auto

Explicit surface, or auto-route

Auto-route: spend / checkout / booking / purchase / payment / cart / Stripe / price / budget / cap → DQL. High-blast irreversible exit without that language (publish, delete, deploy, send-to-prod, memory write) → Sentinel. Unsure → DQL. Explicit mode wins. RV / PLV are not on this path.

Camera mandate: do not put the overshoot in proposed_action or reasoning (for example, do not write “price is above the cap”). The verifier has to find the mismatch.

Envelope (always this shape):

{
  "verdict": "ALLOW",
  "execute": true,
  "objections": [],
  "receipt_id": "dql_…",
  "surface": "dql",
  "axes": [],
  "recommendation": "execute"
}

execute is true only on ALLOW. REVIEW, UNCERTAIN, BLOCK, timeouts, HTTP 402/4xx/5xx, and missing keys return execute: false. Fail-closed is soft at the protocol layer — the tool does not hard-stop the host. Replan is a new call (new receipt).

verify_claim

Verify any claim or AI-generated reasoning via RV (POST /v1/check). Unchanged.

Parameter

Type

Default

Description

claim

string

(required)

The text to verify

stakeLevel

low / medium / high / critical

medium

Risk level — higher stakes trigger deeper verification

domain

financial / medical / legal / code / general

general

Domain context for specialized verification

speed

fast / standard / deep

standard

Verification depth

check_agent_score

Look up an agent's composite trust score on the ERC-8004 registry.

Parameter

Type

Description

agentId

string

Agent ID to look up

domain

string

Optional domain filter

verify_trade

Optional pre-execution gate for trading agents (Sentinel → RV). Not the default verify_decision path. See VERIFY_TRADE.md.

Configuration

Environment Variable

Default

Description

DQL_API_KEY

(none)

DQL key (dqlk_…) for the default verify_decision path. Alias: THOUGHTPROOF_DQL_KEY

SENTINEL_API_KEY

(none)

Optional. Required only when mode=sentinel or auto-route picks Sentinel. Fallback: THOUGHTPROOF_API_KEY as X-Sentinel-Key

DQL_SANDBOX

(off)

Set to 1 to send sandbox: true on DQL calls (local/dev only)

THOUGHTPROOF_API_KEY

(none)

Operator key for verify_claim / verify_trade / Sentinel fallback

THOUGHTPROOF_BASE_URL

https://api.thoughtproof.ai

RV API base URL (verify_claim)

A missing Sentinel key returns execute: false with “Sentinel key not configured” — it does not silently call DQL.

Development

git clone https://github.com/ThoughtProof/thoughtproof-mcp.git
cd thoughtproof-mcp
npm install
npm run build
npm test
npm run dev          # Run with tsx (hot reload)
npm run inspect      # Test with MCP Inspector

For local MCP clients, point command at node and args at dist/index.js after npm run build.

Related

License

MIT — ThoughtProof

Available Tools

1 tool
verify_reasoningB

Verify a decision or reasoning claim using ThoughtProof adversarial multi-model critique. Returns a verdict (ALLOW/HOLD/UNCERTAIN/DISSENT), confidence score, and up to 3 key objections.

ParametersJSON Schema
NameRequiredDescriptionDefault
claimYesThe decision or reasoning to verify
stakeLevelNoStakes of the decision — affects confidence thresholdmedium
domainNoDomain context for the verificationgeneral

TDQS

B3.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions the verification methodology ('ThoughtProof adversarial multi-model critique') and output details (verdict types, confidence score, objections), which adds useful context beyond basic functionality. However, it doesn't cover aspects like rate limits, authentication needs, or error handling, leaving gaps in behavioral understanding.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise and front-loaded, stating the core purpose in the first clause. The second part efficiently lists the return values. There's no wasted text, but it could be slightly more structured (e.g., separating purpose from output details with a colon or bullet points) to enhance readability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (3 parameters, no output schema, no annotations), the description is somewhat complete but has gaps. It explains the purpose and output format, which is helpful, but lacks usage guidelines and full behavioral context. Without an output schema, it should ideally detail return values more thoroughly, though it does list them briefly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the input schema already documents all parameters thoroughly. The description doesn't add any additional meaning or context for the parameters beyond what the schema provides. This meets the baseline score of 3 when the schema handles parameter documentation effectively.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Verify a decision or reasoning claim using ThoughtProof adversarial multi-model critique.' It specifies the verb ('verify') and resource ('decision or reasoning claim'), and mentions the methodology. However, with no sibling tools, it doesn't need to differentiate from alternatives, so it doesn't reach the highest score of 5.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives, prerequisites, or exclusions. It only states what the tool does, not the context for its application. This lack of usage instructions limits its helpfulness for an AI agent.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 1 tool updatev0.1.0
    • First observedverify_reasoning
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

B3.3/5.0
Disambiguation5/5

With only one tool, there is no possibility of ambiguity or overlap between tools. The tool 'verify_reasoning' has a clear, distinct purpose focused on adversarial critique of reasoning claims.

Naming Consistency5/5

A single tool inherently has perfect naming consistency, as there are no other tools to compare it against. The name 'verify_reasoning' follows a clear verb_noun pattern and is descriptive.

Tool Count2/5

One tool is too few for a server's typical scope, as it severely limits functionality and suggests an incomplete or overly narrow implementation. This is borderline for the 'too few' category, making it inappropriate for most practical agent use cases.

Completeness2/5

The server's domain appears to be reasoning verification, but with only one tool, there are significant gaps. For example, there are no tools to list, create, or manage reasoning claims, or to handle different aspects of the critique process, making the surface incomplete for the inferred purpose.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Thoughtproof-mcp

    ThoughtProof
    A
    license
    Not graded
    quality
    C
    maintenance
    Adversarial multi-model reasoning verification for AI agents. Claude, Grok, and DeepSeek challenge each decision — returns ALLOW or HOLD with JWKS-signed attestation. x402-gated on Base.
    5
    MIT

  • agentshield-mcp

    yeick010
    F
    license
    A
    quality
    Not graded
    maintenance
    Trust infrastructure for AI agents on Base. DEX Spread Oracle (live Uniswap V3 prices), on-chain escrow, insurance pool, and collective knowledge base. 7 smart contracts. Pay-per-query via x402 micropayments in USDC.
    6
    -

  • SENTINEL

    teodorofodocrispin-cmyk
    F
    license
    A
    quality
    C
    maintenance
    Agent-Level Transaction Safety Oracle. Before an AI agent signs a blockchain transaction, it returns a SAFE/UNSAFE verdict with a SENTINEL Score (AAA-D) and risk flags. Pay-per-call $0.005 USDC via x402 on Base.
    1
    -

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ThoughtProof/thoughtproof-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server