# Vouchwell (npm · vouchwell) Read, verify and export a tamper-evident log of what an AI agent did. Read-only. - Trust score: 92/100 (high trust) - Change this week: +3 - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-11-05 ## Components - npm · `vouchwell`: 92/100 (this document), [markdown](https://verifymcp.io/servers/vouchwell-vouchwell/vouchwell.md), [page](https://verifymcp.io/servers/vouchwell-vouchwell/vouchwell) ## Channel facts - Registry: `npm` - Package: `vouchwell` - Version: `0.8.0` - 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 2026-11-05. - **Supply Chain Security**: 100/100 - No malware found by supply-chain analysis. - No known CVEs affecting this package version or its production dependencies. - No install/post-install scripts declared. - 0 of 2 dependencies flagged as unhealthy. - **Provenance & Transparency**: 100/100 - Source repository is publicly reachable at the declared URL. - Cryptographically verified build provenance (signed, bound to Vouchwell/vouchwell). - Clear OSI-approved license (Apache-2.0). - Actively maintained (last published 22 days ago). - Publishes a security disclosure policy (SECURITY.md). - **Schema Quality & AI Usability**: 75/100 - AI-judged instruction clarity (good). - Tool/resource definitions use about 557 tokens (~111/item across 5 items; 5 tools + 0 resources), lean. - Usage-examples check failed: none of the tools include examples. - **Stability & Change Management**: 73/100 - Stability observed for 22 of 30 days with no destabilising changes; credit accrues until the full window elapses. - **Tool Coverage**: 100/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 100% of tool parameters carry a description. - **Tool Safety**: 100/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation. - An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 100/100 - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28. ## Install ### How do I install the Vouchwell MCP server? Vouchwell runs locally as an npm package, launched with npx -y vouchwell. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add vouchwell-vouchwell -- npx -y vouchwell ``` ### Cursor ```json { "mcpServers": { "vouchwell-vouchwell": { "command": "npx", "args": [ "-y", "vouchwell" ] } } } ``` ### VS Code ```json { "servers": { "vouchwell-vouchwell": { "command": "npx", "args": [ "-y", "vouchwell" ] } } } ``` ### Codex ```bash codex mcp add vouchwell-vouchwell -- npx -y vouchwell ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "vouchwell-vouchwell": { "type": "local", "command": [ "npx", "-y", "vouchwell" ], "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add vouchwell-vouchwell --command npx --arg -y --arg vouchwell ``` ### Hermes ```yaml mcp_servers: vouchwell-vouchwell: command: "npx" args: ["-y", "vouchwell"] ``` ### Netclaw ```json { "McpServers": { "vouchwell-vouchwell": { "Transport": "stdio", "Command": "npx", "Arguments": [ "-y", "vouchwell" ] } } } ``` ### Vellum ```bash assistant mcp add vouchwell-vouchwell -t stdio -c npx -a -y vouchwell ``` ### Other ```json { "mcpServers": { "vouchwell-vouchwell": { "command": "npx", "args": [ "-y", "vouchwell" ] } } } ``` ## 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. ### 2026-11-05 (score 92, +1) No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-11-03 (score 91, +1) No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-11-01 (score 90, +1) - [functional] Package version: 0.7.0 → 0.8.0 ### 2026-10-29 (score 89, +1) No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-10-27 (score 88, +1) No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-10-25 (score 87, +1) No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-10-23 (score 86, +1) No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-10-21 (score 85, +3) - [functional improvement] Stability: unverified → 0.23 ## MCP tools (5) ### `log_entries` (~161 tokens) Recent receipts The most recent receipts in the Vouchwell log: what was called, when, and whether the policy allowed, refused or escalated it. Filters narrow it to refusals, calls monitor mode would have blocked, calls that never finished, a tool, or a session. Input parameters: - `denied` (boolean): Only calls the policy refused or escalated. - `session` (string): Only this session. - `tail` (integer): How many of the latest matching receipts to return (default 20). - `target` (string): Only tools whose name contains this text. - `unfinished` (boolean): Only calls that were authorised and sent but never reported a result. - `would_block` (boolean): Only calls monitor mode let through that the policy would have blocked. ### `verify_log` (~52 tokens) Verify the log Check every signature, hash-chain link and signed checkpoint in the Vouchwell log. ok: false means the log was altered after it was written: report it as evidence of tampering, never as something to repair. ### `log_stats` (~41 tokens) Log totals Totals for the Vouchwell log: receipts, allowed, refused, escalated, tool errors, what monitor mode would have stopped, and the busiest tools. ### `check_bundle` (~99 tokens) Check an evidence bundle Verify a Vouchwell evidence bundle file using nothing but the file itself. Pin the signer with log_keys (kid=publicKey, as the log owner published them) to also prove who signed it. Input parameters: - `log_keys` (object): Trusted signing keys, kid → base64url public key. Without them, the bundle is checked against the keys it carries. - `path` (string, required): The bundle file, relative to the working directory or absolute. ### `export_evidence` (~131 tokens) Write an evidence bundle Write the log as a new evidence bundle file anyone can verify offline with `vw check`; the log itself is not changed. It holds no payloads and no salts; argument previews, if the log records them, are included. Refuses to overwrite an existing file. Input parameters: - `path` (string): Where to write it, relative to the working directory (default: vouchwell--