# ThoughtProof (npm · thoughtproof-mcp)

Pre-action gate: verify before pay/trade/write/deploy. execute=true only on ALLOW.

- Trust score: 64/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-20

## Components

- npm · `thoughtproof-mcp`: 64/100 (this document), [markdown](https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp.md), [page](https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp)

## Channel facts

- Registry: `npm`
- Package: `thoughtproof-mcp`
- Version: `0.3.2`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-20.

- **Supply Chain Security**: 98/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 30 of 96 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 1 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 64/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1066 tokens (~213/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 10/100
  - Stability observed for 3 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 60/100
  - 40% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add thoughtproof-thoughtproof-mcp -- npx -y thoughtproof-mcp
```

### Codex

```bash
codex mcp add thoughtproof-thoughtproof-mcp -- npx -y thoughtproof-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "thoughtproof-thoughtproof-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "thoughtproof-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add thoughtproof-thoughtproof-mcp --command npx --arg -y --arg thoughtproof-mcp
```

### Hermes

```yaml
mcp_servers:
  thoughtproof-thoughtproof-mcp:
    command: "npx"
    args: ["-y", "thoughtproof-mcp"]
```

### Other

```json
{
  "mcpServers": {
    "thoughtproof-thoughtproof-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "thoughtproof-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-20 (score 64, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-19 (score 63, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-08-18 (score 48, +1)

- [security improvement] Malware scan: unverified → pass
- [functional regression] Schema quality: 169 → 213
- [functional improvement] Tool coverage: 25% → 40%
- [functional improvement] Stability: unverified → 0.03
- [functional] Package version: 0.3.1 → 0.3.2

### 2026-08-17 (score 47)

First indexed and scored.

## MCP tools (5)

### `verify_decision` (~350 tokens)

ThoughtProof pre-action verification gate: verify before the agent pays, trades, writes, or deploys. Call this BEFORE any consequential tool (payment, trade, transfer, publish, delete, deploy, irreversible write). Returns { verdict: ALLOW|BLOCK|UNCERTAIN|…, execute: boolean, objections, receipt_id, surface, axes?, recommendation }. execute is true ONLY on ALLOW — on any other result do NOT execute (soft fail-closed; host must honor execute=false). Routes internally to DQL (spend/checkout) or Sentinel (irreversible exit). Camera mandate: you must NOT put the overshoot or constraint violation in proposed_action or reasoning (for example, do not write "price is above the cap"). Put the user goal in mandate, the action you are about to take in proposed_action, and your plan in reasoning — the verifier must find the mismatch. Replan = new call = new receipt. Aliases: verify_decision, verify_before_action, verify_before_act.

Input parameters:

- `context` (string): Optional extra evidence, tool outputs, or prior turns.
- `mandate` (string, required): The user's stated goal or instruction the agent is acting on.
- `mode` (string): Routing override. auto (default) picks DQL for spend/checkout language and Sentinel for high-blast irreversible exits; unsure → DQL. Explicit mode wins.
- `proposed_action` (string, required): What the agent is about to do (pay/trade/write/deploy/…). Do not include the overshoot or constraint violation here.
- `reasoning` (string, required): The agent's own plan or reasoning. Do not include the overshoot; the verifier has to find the mismatch.

### `verify_before_action` (~350 tokens)

ThoughtProof pre-action verification gate: verify before the agent pays, trades, writes, or deploys. Call this BEFORE any consequential tool (payment, trade, transfer, publish, delete, deploy, irreversible write). Returns { verdict: ALLOW|BLOCK|UNCERTAIN|…, execute: boolean, objections, receipt_id, surface, axes?, recommendation }. execute is true ONLY on ALLOW — on any other result do NOT execute (soft fail-closed; host must honor execute=false). Routes internally to DQL (spend/checkout) or Sentinel (irreversible exit). Camera mandate: you must NOT put the overshoot or constraint violation in proposed_action or reasoning (for example, do not write "price is above the cap"). Put the user goal in mandate, the action you are about to take in proposed_action, and your plan in reasoning — the verifier must find the mismatch. Replan = new call = new receipt. Aliases: verify_decision, verify_before_action, verify_before_act.

Input parameters:

- `context` (string): Optional extra evidence, tool outputs, or prior turns.
- `mandate` (string, required): The user's stated goal or instruction the agent is acting on.
- `mode` (string): Routing override. auto (default) picks DQL for spend/checkout language and Sentinel for high-blast irreversible exits; unsure → DQL. Explicit mode wins.
- `proposed_action` (string, required): What the agent is about to do (pay/trade/write/deploy/…). Do not include the overshoot or constraint violation here.
- `reasoning` (string, required): The agent's own plan or reasoning. Do not include the overshoot; the verifier has to find the mismatch.

### `verify_claim` (~116 tokens)

Input parameters:

- `claim` (string, required): The claim or AI-generated reasoning to verify. Can be a statement, decision, analysis, or any text that needs adversarial verification.
- `domain` (string): Domain context for specialized verification. Default: general
- `speed` (string): Verification depth. fast=$0.008 (2 models), standard=$0.02 (4 models), deep=$0.08 (5+ models). Default: standard
- `stakeLevel` (string): Risk level — higher stakes trigger more thorough verification. Default: medium

### `check_agent_score` (~43 tokens)

Input parameters:

- `agentId` (string, required): The agent ID to look up (e.g. 'agent_abc123xyz')
- `domain` (string): Optional domain filter for the score

### `verify_trade` (~207 tokens)

Input parameters:

- `action` (string, required): The action the agent intends to execute, e.g. 'open 5x long BTC, 8000 USDC margin' or 'swap 10 ETH for USDC'.
- `reasoning` (string, required): The full reasoning chain that produced the decision — the actual chain of thought, not a summary. This is what ThoughtProof verifies.
- `situation` (string): Optional market/context snapshot WITHOUT the chosen action. Lets the adversarial panel form an independent view before seeing the decision (stronger verification).
- `stakeLevel` (string): Stake level — drives the verdict threshold. Higher stake demands higher reasoning soundness to ALLOW: the SAME decision can ALLOW at low stake and BLOCK/UNCERTAIN at critical stake. Default 'high' (s…
- `thesis` (string, required): The agent's one-line decisive rationale for the action.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp#diagnostics

## Score history

- 2026-08-20: 64
- 2026-08-19: 63
- 2026-08-18: 48
- 2026-08-17: 47

## Links

- npm package: https://www.npmjs.com/package/thoughtproof-mcp
- Socket report: https://socket.dev/npm/package/thoughtproof-mcp
- Repository: https://github.com/ThoughtProof/thoughtproof-mcp
- Website: https://thoughtproof.ai/
- Changelog RSS feed: https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp.json
- HTML version of this page: https://verifymcp.io/servers/thoughtproof-thoughtproof-mcp/thoughtproof-mcp