
ThoughtProof
io.github.ThoughtProofPublished by GitHub account ThoughtProofv0.3.2Updated Nov 10, 2026
Pre-action gate: verify before pay/trade/write/deploy. execute=true only on ALLOW.
The publisher's own server: Proof Holdings
Overview
Local stdio MCP server that lets an AI agent verify a proposed action before paying, trading, writing or deploying, returning a fail-closed execute flag.
- What you can do
- Verify a proposed action before execution: submit mandate, proposed action and reasoning to get a verdict, objections and receipt, with routing to spend or irreversible-exit checks
- Verify claims and AI-generated reasoning at a chosen stake level, domain and depth
- Look up an agent's composite trust score on the ERC-8004 registry
- Run an optional pre-execution gate for trading agents
- Return a fixed envelope where execute is true only on ALLOW
- When to use it
- Use it when an agent is about to take a consequential step — spend money, check out, publish, delete, deploy or write memory — and you want an independent check before it acts. Also useful for fact-checking claims or reasoning, or looking up agent trust scores.
- Requirements
- Node.js 18+ and a local stdio MCP client such as Claude Desktop, Cursor, Windsurf or Cline; run via npx. A DQL_API_KEY is needed for the default verify_decision path; SENTINEL_API_KEY is needed when mode=sentinel or auto-route picks Sentinel; THOUGHTPROOF_API_KEY is an operator key for verify_claim, verify_trade and Sentinel fallback. Keys are obtained from the provider's pricing page, and network access to the provider API is required.
Installation
In SourceWeft
- Open ThoughtProof in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
thoughtproof-mcp
[npm version] [CI] [HOL Guard Scanner] [HOL Verified] [License: MIT]
thoughtproof-mcp — local stdio. Hero tool verify_decision (DQL spend / Sentinel irreversible exit). execute is true only on ALLOW. Sentinel ALLOW must match a verified signed export.
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
Install with npx -y [email protected]. Works with Claude Desktop, Cursor, Windsurf, Cline, and other local stdio MCP clients.
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.
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):
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.
check_agent_score
Look up an agent's composite trust score on the ERC-8004 registry.
verify_trade
Optional pre-execution gate for trading agents (Sentinel → RV). Not the default verify_decision path. See VERIFY_TRADE.md.
Configuration
A missing Sentinel key returns execute: false with “Sentinel key not configured” — it does not silently call DQL.
Development
For local MCP clients, point command at node and args at dist/index.js after npm run build.
Security
See SECURITY.md for supported versions, private reporting ([email protected]), and key handling (dqlk_… for MCP hero — never commit real keys).
HOL listing / Guard scanner score is a review baseline only. It is not next-action authorization and not a substitute for fail-closed host behavior (execute: true only on ALLOW).
Related
- ThoughtProof — Decision verification for AI agents
- pot-cli — CLI for reasoning verification
- ERC-8004 — Autonomous Agent Registry
License
MIT — ThoughtProof
Source: README.md at commit cb7a6c6
Tools
0Version history
1- v0.3.2LatestSep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Other Proof integrations
Proof Holdings
holdings.proofOfficial publisher
Proof Holdings MCP server lets an assistant create identity verifications, validate proof tokens and manage delegations through the proof.holdings API.

sentinel-proof
Published by GitHub account 92kareeem
Checks an answer against its sources: numbers and citations verified, wrong ones fixed. No model.

Lean4 formal proof — Atom
org.aimlds
Remote Conduit Atoms service that delivers Lean4 formal proof capability to an assistant per API call.

Ai Inference Proof Escrow Auditor
Published by GitHub account adamgraymon
Remote MCP server for a non-custodial USDC escrow on Base that hires AI providers and audits deliverables before settling or refunding.

LinkProof
dev.workers.bsptimber.linkproof
LinkProof is a remote, pay-per-call MCP server that lets an assistant check URLs, verify citations and find Wayback Machine snapshots before citing them.