elliot-mcp
Elliot is an independent verification MCP server for AI-agent claims and project documentation promises.
Audit README/docs/security/pricing claims against the code: pin a commit, risk-sweep risky promises, record claims with checked quotes or absent searches, and generate HTML/Markdown/JSON reports.
Check “done” claims: snapshot and verify actual file changes, run functional tests in an OS sandbox, verify URL status/text, GitHub PR/CI/commit state, and npm package publication/version timing.
Record undisclosed risky code behavior that docs never mention.
Require independent review of findings before finalizing an audit, including isolated reviewer mode and audit-scoped tokens.
Produce honesty reports from receipts and verify Ed25519-signed receipts/attestations.
Allows auditing GitHub pull requests by checking claims in the PR description against the code and commenting with proof of verification or falsehood.
Quick start
In the repository you want to check:
npx -y github:EmanueleCell4/elliot-mcp initRestart your agent, then type:
/elliot-auditA minute or three later the report is in elliot-report/<date>-<commit>/ at the root of the repo: report.html, report.md and claims.json. There is nothing to open, no dashboard and no account.
Requirements: Node 20 or newer, git, and Claude Code, Cursor or Codex.
As a Claude Code plugin
Inside Claude Code:
/plugin marketplace add EmanueleCell4/elliot-mcp
/plugin install elliot@elliot-mcpThe first command registers Elliot's plugin marketplace; the second installs Elliot. After installation, run /elliot:audit in your project. This is an opt-in local tool and does not automatically post reports or install itself in other repositories.
That adds the Elliot tools, a /elliot:audit command, and a skill that makes Claude verify its own "done" claims before reporting them.
On every pull request (GitHub Action)
Elliot reads the pull request description, checks each claim against the code ("fixes X", "adds tests for Y", "no behaviour change"), and comments with the proof. If an agent wrote the PR, this is where its "done" gets checked.
# .github/workflows/elliot.yml
name: Elliot
on:
pull_request:
types: [opened, edited, synchronize]
permissions:
contents: read
pull-requests: write
jobs:
check:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
with:
fetch-depth: 0
- uses: EmanueleCell4/elliot-mcp@v0.7.3
with:
mode: strict
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}The comment looks like this:
Verdict | Claim | Proof |
❌ FALSE | Adds unit tests for the comment builder in test/comment.test.js | searched for |
✅ VERIFIED | The badge is only offered when the audit has no findings |
|
It runs Claude Code with your API key. One check usually takes 1–2 minutes. Nothing is sent anywhere except Anthropic's API and GitHub.
mode: strict (shown above, available from v0.7.0) fails the check for findings, missing evidence, skipped audits and inconclusive results, so a merge can be blocked on it: add the job as a required status check in your branch protection rules. Drop mode to get the non-blocking report instead. A ready-to-copy workflow is in examples/workflows/elliot-strict.yml. Do not enable enforcement on forked PRs without reviewing credential exposure and runner isolation.
README badge
When every claim Elliot checks holds up, the report includes a badge you can paste into your README:
Related MCP server: agentseed
Why
Coding agents say "Done ✅" when the work isn't done. They delete failing tests, skip checks, describe security that doesn't exist, and write READMEs that promise what the code never does. CI stays green because tests check the code the agent wrote, not the promises the product makes.
I found this out the hard way. My own legal-tech app had a security page promising role-based access enforced in the database. The data layer bypassed it by design, and every permission check was a function that ignored its input. The agent had told me it was done.
The usual fix is to ask a second AI to review the first one. That gives you another opinion from the same kind of system that made the mistake. In the first automatic audit I ran, the agent doing the checking cited line numbers that don't exist in 4 of 17 submissions.
Elliot works the other way around: the agent proposes, Elliot verifies. Every quote must really be in the file, at that line, at that commit: the audit fingerprints the tree when it starts (tracked changes and untracked files included) and refuses evidence once anything changes, and symlinks that lead outside the repository are not evidence. Every "this doesn't exist" is searched by Elliot itself. A fresh reviewer then tries to prove each finding wrong before it reaches the report.
What it found
Elliot 0.3 ran unattended inside Claude Code on 9 popular open-source tools from the AI coding-agent ecosystem (October 2026). A separate reviewer that had not seen the audits then checked every finding against the code.
Findings in the reports | Confirmed correct | Left unconfirmed by the reviewer | Wrong | Time per repo |
13 | 12 | 1 | 0 | 1 to 3 minutes |
"Wrong" means a factual error: the code does not show what the finding says.
Some of what it caught:
A tool whose README says "No Telemetry: no data collection or tracking" ships analytics that are on by default, with consent pre-set to "yes". (public issue)
An app where logging in turns telemetry back on after the user switched it off, then identifies the user by email. (public issue)
A security page that calls a script sandbox "read-only" while the code can give scripts file writes and MCP tools. Confirmed by the project's triage bot, with a docs fix sent upstream. (issue, PR)
Docs saying an API key is stored encrypted while the code writes it as plain JSON.
A popular agent tool that sends users' full AI prompts and responses to a third-party error tracker by default, under an option named "anonymousTelemetry". Reported privately to the maintainers; details will be published after a fix.
What it still misses: on the same 9 repos it found 6 of the 10 serious issues I had found by hand beforehand (a small, hand-picked set, so read it as a rough coverage figure, not a recall estimate). The most common miss is a missing authorization check (for example, a download endpoint that doesn't check who is asking). Elliot doesn't look for those yet; see the roadmap.
Benchmark: Claude with Elliot vs Claude alone
Eight widely used SDKs, 18 broken promises planted by flipping a value behind a documented default, two runs per condition, same model and instructions. A blind judge checked every finding that was not a planted bug.
Claude alone | Claude + Elliot | |
Planted bugs found (2 runs) | 28 / 36 (78%) | 30 / 36 (83%) |
False alarms (judged not real) | 21% of findings | 6% of findings |
Real unplanted problems found | 8 | 5 |
Elliot raises about a third as many false alarms and finds slightly more planted bugs; Claude alone casts a wider net. When the two blind judges (Claude and Codex) disagreed, reading the code showed Codex had misstated the code 9 times: a second model is not a check unless something re-reads the source. Method, limits and all raw outputs: benchmark/planted-promises.
Benchmark: can anyone re-check what the agent says it verified?
56 real bug fixes (28 tasks, Opus and Haiku). Every fix was correct and no agent claimed "done" falsely. But in 19 of 56 runs (34%) the check the agent said it ran could not be re-run by anyone else: a deleted scratch script, a check described in prose, a path on the agent's own machine. The work was right; the evidence was gone. benchmark/done-claims
How it works
Risk sweep (Elliot). Before the agent reads anything, Elliot scans the docs for risky promises (telemetry, privacy, encryption, access control, sandboxes, "never", "by default") and the code for signals that often contradict them: analytics SDKs, permission bypasses such as
--dangerously-skip-permissions, privileged database clients, secrets written to disk. The audit cannot finish until every one is addressed or skipped with a reason.Read and propose (your agent). Your agent traces each claim into the code and proposes a verdict, quoting the docs and the code.
Verify the proof (Elliot). Each quote is matched against the file at the pinned commit, within two lines of where the agent said it is. Each "absent" claim is searched by Elliot with
git grep. Citations that don't check out are rejected, and the agent has to fix them.Materiality test. A FALSE or PARTIAL verdict must say what a user who trusts the docs would get wrong. Omissions, naming details and tiny delays become notes, not findings. Changelogs, roadmaps and internal developer docs are rejected as sources, because they are not promises to users.
Independent review. A fresh subagent that never saw the audit tries to prove every finding wrong, and confirms, downgrades or retracts it. Elliot will not write the report until every finding has been reviewed.
Report. HTML, Markdown and JSON in
elliot-report/, plus a tamper-evident receipt for every check in.elliot/receipts.jsonl.
If you wrote code in the same session, the protocol tells the agent to run the audit in a fresh subagent, so the auditor is not the author.
Example report
The full report is in examples/lex-ai-report.html. It was produced by an earlier version, before the independent review step existed.
Commands
Set-up: init
npx -y github:EmanueleCell4/elliot-mcp init # every agent it detects
npx -y github:EmanueleCell4/elliot-mcp init --claude # only Claude Code
npx -y github:EmanueleCell4/elliot-mcp init --cursor # only Cursor
npx -y github:EmanueleCell4/elliot-mcp init --codex # only Codex
npx -y github:EmanueleCell4/elliot-mcp init --all # all threeinit is idempotent: running it twice changes nothing, and it never overwrites a command file you edited. It writes:
Agent | MCP server registered in | Command added |
Claude Code |
|
|
Cursor |
|
|
Codex |
|
|
Audit a repo: /elliot-audit
Run it in your agent from the repository root. In Claude Code and Codex, anything you type after the command is passed to the agent as extra instructions, for example /elliot-audit focus on the security page.
Check a "done" claim
The same server also checks individual claims while the agent works. It ships instructions telling the agent to use them before it says "done".
The agent says | Tool | What Elliot looks at |
"I fixed / cleaned / refactored the code" |
| the content hash of every file before and after the task |
"The behavior works" |
| runs the requested test command in a temporary OS sandbox; only exit code 0 is VERIFIED |
"It's deployed / live" |
| the HTTP status and the text really served at the URL |
"PR merged", "CI is green", "pushed" |
| the GitHub API: merge state, check runs, whether the commit exists |
"Published to npm" |
| the npm registry: version exists, published after a given time |
"What did my agent really do today?" |
| your receipts: claims checked, how many held up |
These checks answer VERIFIED, NOT_VERIFIED or INCONCLUSIVE. Inconclusive means Elliot could not confirm, for example because of a GitHub rate limit. It never guesses.
All MCP tools
Tool | Used by | What it does |
| the agent | Pins the commit, lists the files that make claims, runs the risk sweep, returns the protocol. |
| the agent | Records one claim with a verdict. Checks the source quote, every evidence quote and every absence search. |
| the agent | Records risky code behaviour the docs never mention, with checked code quotes. |
| the agent | Returns a self-contained prompt for a fresh reviewer subagent. |
| the reviewer | Confirms, downgrades or retracts a finding. A retraction needs a code quote. |
| the agent | Refuses to finish while risk-sweep items or reviews are missing, then writes the report. |
| the agent | Fingerprints the working tree before a task. |
| the agent | Compares the tree with a snapshot, or with HEAD. |
| the agent | Runs a command and arguments against a disposable copy, with writes confined to the copy and network blocked. |
| anyone | Checks an Ed25519 signature and receipt digest, optionally against a separately trusted public key. |
| the agent | Checks status and expected text at a URL. |
| the agent | Checks |
| the agent | Checks a package version on the registry. |
| you | Summarises recent receipts. |
Verdicts in an audit
Verdict | Meaning |
| The code implements the claim. A small gap that misleads nobody is kept as a note. |
| Implemented, with a gap that would make a user who relies on it do something wrong. |
| The code contradicts the claim, or the feature, command, flag or option does not exist. |
| Depends on a deployment, a contract or a service outside the repo. |
| Risky behaviour in the code that the docs never mention. |
Manual installation
Use this if you'd rather not run init.
Claude Code
claude mcp add elliot -- npx -y github:EmanueleCell4/elliot-mcpCodex (~/.codex/config.toml)
[mcp_servers.elliot]
command = "npx"
args = ["-y", "github:EmanueleCell4/elliot-mcp"]Cursor (.cursor/mcp.json) and any other MCP client
{ "mcpServers": { "elliot": { "command": "npx", "args": ["-y", "github:EmanueleCell4/elliot-mcp"] } } }Then add the command yourself, or ask your agent to "run an Elliot claim audit of this repository". For agents that support skills, also add skills/elliot/SKILL.md.
Configuration
All settings are optional environment variables.
Variable | Default | What it does |
| unset | Used by |
|
| Where receipts and audit state are stored. |
|
| GitHub API base URL, for GitHub Enterprise or tests. |
|
| Registry used by |
| unset | Path to an Ed25519 PKCS#8 private key. When set, every saved receipt is signed. Keep it outside the repository. |
| public-key fingerprint | Human-readable signer label included in signed receipts. The label is not an identity proof by itself. |
| unset | Set to |
| unset | Enables isolated review mode; configure the same non-empty secret on auditor and reviewer server processes. |
Network access and data
Elliot has no telemetry and no server of its own. What it touches:
Audit (
/elliot-audit,/elliot:audit): reads files in the repository and runsgitlocally. It makes no network calls.verify_url: sends one GET request to the URL the agent passes in. Loopback, private-network and link-local addresses (including cloud metadata) are refused on every redirect hop unless you setELLIOT_ALLOW_PRIVATE_URLS=1; one 15-second deadline covers the whole request and the body stops at 2 MiB.verify_github: callsapi.github.com(orELLIOT_GITHUB_API). If a GitHub token is set (theGITHUB_TOKENenv var, or the plugin's optionalgithub_tokensetting), it is sent only to that API.verify_npm: callsregistry.npmjs.org(orELLIOT_NPM_REGISTRY).GitHub Action: runs Claude Code in your CI with your
ANTHROPIC_API_KEY, reads the PR description and code, and posts one comment through the GitHub API.On disk: receipts in
.elliot/(orELLIOT_DIR) and reports inelliot-report/, both inside your repository.
The Claude Code plugin runs the server bundled in this repository (node src/index.js). Claude Code installs its two npm dependencies from package-lock.json when you install the plugin.
Output
your-repo/
├── elliot-report/
│ └── 2026-10-22-dd84474/
│ ├── report.html # open in a browser
│ ├── report.md # paste into an issue or PR
│ └── claims.json # everything, for scripts and CI
└── .elliot/
├── receipts.jsonl # one tamper-evident receipt per check
└── audits/ # audit stateAdd .elliot/ to .gitignore. Commit elliot-report/ if you want the report to live with the code.
A receipt looks like this:
{
"elliot": "0.3.0",
"id": "rcpt_3f9a1c2b7d10",
"check": "claim",
"claim": "Role-based access is enforced at the database layer through row-level security.",
"target": "lex-ai@dd84474",
"verdict": "FALSE",
"reason": "All data access uses the service-role client, which bypasses RLS.",
"evidence": { "source": { "file": "app/security/page.tsx", "line_start": 14 }, "evidence": [ { "file": "lib/supabase/index.ts", "line_start": 10 } ] },
"checked_at": "2026-10-21T19:42:11.000Z",
"checker": "local",
"digest": "sha256:…"
}The digest detects changes to a receipt. For third-party verification, configure a private key outside the repository:
openssl genpkey -algorithm Ed25519 -out "$HOME/.elliot-signing-key.pem"
chmod 600 "$HOME/.elliot-signing-key.pem"
export ELLIOT_SIGNING_KEY="$HOME/.elliot-signing-key.pem"
export ELLIOT_SIGNER_ID="acme-ci"Saved receipts then include the Ed25519 public key, its SHA-256 fingerprint, signer label and signature. A verifier checks the digest and signature with verify_attestation; valid is true only when a separately trusted public key is supplied and matches. Without that key, the verifier reports cryptographic validity while marking signer identity untrusted. The embedded key proves which key made the signature, not that the key belongs to a particular person or organization. Publish or distribute the trusted public key through a channel your verifiers already trust. Protect and back up the private key outside the repo. Receipts created without ELLIOT_SIGNING_KEY remain unsigned for compatibility.
Isolated review
For enforced reviewer separation, run Elliot twice with the same ELLIOT_DIR and a private ELLIOT_REVIEW_TOKEN (generate one with openssl rand -hex 32). The auditor process uses ELLIOT_ROLE=auditor; the independent reviewer process uses ELLIOT_ROLE=reviewer. The auditor server omits elliot_record_review; the reviewer server exposes only that tool. elliot_review_packet gives the reviewer the finding, checked quotes, impact and an audit-scoped capability, without repository access or the auditor's explanation. Elliot encrypts the capability in audit state, checks its hash and the reviewer server credential on submission, and rejects cross-audit reuse and duplicate decisions. Keep the shared ELLIOT_REVIEW_TOKEN in the two host configurations; only the single-audit capability travels in the review packet.
{
"mcpServers": {
"elliot-auditor": {
"command": "node", "args": ["/path/to/elliot-mcp/src/index.js"],
"env": { "ELLIOT_ROLE": "auditor", "ELLIOT_REVIEW_TOKEN": "<same-random-secret>", "ELLIOT_DIR": "/path/to/project/.elliot" }
},
"elliot-reviewer": {
"command": "node", "args": ["/path/to/elliot-mcp/src/index.js"],
"env": { "ELLIOT_ROLE": "reviewer", "ELLIOT_REVIEW_TOKEN": "<same-random-secret>", "ELLIOT_DIR": "/path/to/project/.elliot" }
}
}
}The host must give the reviewer only the packet and the reviewer MCP connection. The default unconfigured mode retains the original combined tool surface for existing MCP clients. Separate processes and tool permissions enforce the boundary; deployments where the auditor can read reviewer host configuration or impersonate its MCP connection do not provide a separate trust domain.
Functional outcome checks
Call verify_outcome with an executable and argument array, for example command: "npm", args: ["test"]. Elliot copies the working tree to a temporary directory, then runs the command in the OS sandbox (sandbox-exec on macOS or bwrap on Linux), with network access blocked and writes limited to the temporary copy. A zero exit code is VERIFIED; a failing check is NOT_VERIFIED. If the OS sandbox is unavailable, the result is INCONCLUSIVE and the command is not run. Dependencies already present in node_modules are copied into the temporary tree. No shell expansion is performed; pass flags as separate arguments.
FAQ
Isn't this just an LLM judging an LLM?
No. The agent proposes verdicts, but every piece of evidence is checked deterministically: quotes against files at a pinned commit, absences with Elliot's own git grep. A second agent that never saw the audit reviews every finding, and a retraction has to be backed by a code quote that Elliot also checks.
Does it send my code anywhere?
Elliot itself makes no network calls during an audit. It reads files and runs git locally. Your agent reads the code as it normally does, with the model you already use. The verify_url, verify_github and verify_npm checks call those services only when the agent uses them.
Does it change my code?
No. It writes only .elliot/ and elliot-report/, and init writes the config files listed above.
What does it cost? Nothing. Elliot is MIT-licensed and the audit runs on the agent and model you already pay for. An audit of a medium-sized repo took 1 to 3 minutes in our tests.
Can it audit private repos?
Yes. It runs where your code is. Only verify_github needs GITHUB_TOKEN for private repos.
Can I trust a PASS?
Treat a clean report as "nothing found among the claims checked", not as a certificate. The claim audit is a static reading of the code; it does not run your app (use verify_outcome for that). Authorization checks are covered only where sibling handlers check the caller and one does not; ownership checks are not covered yet.
I found a false positive.
Please open an issue with the report's claims.json entry. False positives are the number Elliot is built to keep at zero.
Roadmap
File, URL, GitHub and npm checks with local receipts
/elliot-audit: claim audit with checked citations, report inelliot-report/Risk sweep, undisclosed behaviours, materiality test and independent review
Authorization gaps: handlers that skip the auth check their sibling handlers perform
Ownership checks: handlers that check the caller is logged in but not that they own the record
Run functional outcome checks in an OS sandbox
Hooks that run Elliot automatically when the agent stops (Claude Code, Codex)
A GitHub Action that checks the claims in each pull request description
Ed25519 signed receipts with offline verification
Development
git clone https://github.com/EmanueleCell4/elliot-mcp && cd elliot-mcp
npm install
npm test # 35 tests, including end-to-end runs that drive the server over stdio with the MCP clientThe server is plain Node (ES modules) with two dependencies: the MCP SDK and zod. The main pieces:
File | What it does |
| MCP server, tool definitions, |
| Audit state, citation checks, review gate, verdict resolution |
| Risk sweep: risky doc promises and code signals |
| The protocol the agent follows |
| HTML and Markdown reports |
| Registers Elliot in Claude Code, Cursor and Codex |
|
|
Issues, false positives and stories of agents telling you they were done are very welcome.
License
MIT © 2026 Emanuele Cella
Available Tools
14 toolselliot_audit_finishFinish the audit and write the reportA
Close the audit and write report.html, report.md and claims.json into elliot-report/-/ at the repository root.
Inputs: audit_id (required), out_dir, skipped.
elliot_audit_startStart a claim auditA
Start auditing whether this repository does what its README, docs and security/pricing pages say. Pins the commit, lists the files that make claims, and returns the protocol to follow.
Inputs: path, sources.
elliot_record_behaviorRecord undisclosed risky behaviorA
Record something risky the code does that the docs never mention (for example, it sends usage events, stores a key in plain text, or skips permission prompts) and no doc sentence promises otherwise. If a doc sentence does promise otherwise, use elliot_record_claim on that sentence instead. Never attach this to an unrelated sentence.
Inputs: audit_id (required), behavior (required), evidence (required), severity (required), +3 more.
elliot_record_claimRecord one audited claimA
Submit one claim with its verdict and evidence. Elliot checks that the claim's source quote is really in the docs at those lines, that every evidence quote is really in the code at those lines, and runs every 'absent' search itself. If anything does not check out the claim is rejected: fix the citation and submit again.
Inputs: claim (required), source (required), verdict (required), audit_id (required), +5 more.
elliot_record_reviewRecord an independent review decisionA
For the reviewer process: confirm, downgrade or reject one finding. In isolated mode provide the audit-scoped reviewer_token from the evidence packet. Confirm needs evidence you checked yourself (a code quote or an absent search); reject needs a code quote showing the claim holds. All evidence is checked against the files.
Inputs: reason (required), finding (required), outcome (required), audit_id (required), +3 more.
elliot_reportHonesty reportARead-only
Summarise recent receipts: how many claims were checked and how many held up. Show this to the user when they ask what the agent really did.
Inputs: limit.
elliot_review_packetGet the independent review packetA
Returns only findings and checked evidence for review. Give it to a fresh reviewer process. With ELLIOT_REVIEW_TOKEN configured, the auditor endpoint cannot record review decisions; connect the reviewer only to a separate ELLIOT_ROLE=reviewer server.
Inputs: audit_id (required).
elliot_snapshotSnapshot the working treeA
Fingerprint every tracked and untracked file in the git repository before starting a task, so verify_changes can later prove what really changed.
Inputs: path.
verify_attestationVerify a signed receiptARead-only
Verify an Ed25519-signed Elliot receipt and its digest. Returns valid=true only when the signature passes and an explicitly supplied public key matches; without one it reports cryptographic validity but marks signer identity untrusted.
Inputs: receipt (required), public_key.
verify_changesVerify files actually changedARead-only
Check whether files really changed, compared with a snapshot taken before the task (or with HEAD if no snapshot is given). Use before claiming code was fixed, cleaned, refactored or written.
Inputs: path, claim, paths, snapshot_id.
verify_githubVerify a GitHub claimARead-only
Check what GitHub actually says: whether a PR is merged, whether CI checks passed on a PR or commit, or whether a commit/branch was really pushed. Set GITHUB_TOKEN for private repos.
Inputs: repo (required), check (required), pr, ref, +1 more.
verify_npmVerify an npm package was publishedARead-only
Check the npm registry for a package and version. Use before claiming a package or release was published.
Inputs: name (required), claim, version, published_after.
verify_outcomeRun a functional check in isolationARead-only
Run a command and arguments in a disposable repository copy under the host OS sandbox. Writes stay in the copy, network is blocked, and VERIFIED requires exit code zero. For a shell pipeline use command bash with args ["-c", "..."]; Elliot adds set -o pipefail so a failing step is not hidden by the last one. Put every file the check needs in the repository: files elsewhere (such as /tmp scripts) do not exist inside the sandbox. If the OS sandbox cannot start, returns INCONCLUSIVE without running the command.
Inputs: command (required), args, path, claim, +1 more.
verify_urlVerify a URL is liveARead-only
Check that a URL (a deploy, a preview, a page) actually responds with the expected status and, optionally, contains expected text. Use before claiming something is deployed or live.
Inputs: url (required), claim, expect_text, expect_status.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 14 tools
Most tools target clearly different resources: verify_changes (file diffs), verify_url (HTTP), verify_outcome (sandboxed command), verify_github, verify_npm, verify_attestation are unambiguous. Weaker spots are the report-family (elliot_report summarises receipts, elliot_review_packet packages findings, elliot_audit_finish writes report files) and elliot_record_claim vs elliot_record_behavior, which are distinguished only by careful reading of the descriptions.
There is a coherent, if dual, convention: elliot_* for audit-state tools and verify_* for external-evidence checks, each with readable verb/noun content. Minor deviations exist (elliot_audit_start / elliot_audit_finish use noun_verb ordering while elliot_record_claim, elliot_record_behavior, elliot_record_review use verb_noun).
14 tools sits comfortably in the reasonable band, and each verify_* variant covers a genuinely different evidence source rather than padding. The only mild bloat is the three report/output tools, though each serves a different consumer (user summary, reviewer process, on-disk artifacts).
The audit lifecycle is well covered end to end: snapshot, audit_start, claim/behavior recording, review_packet, record_review, audit_finish, report. Gaps are minor: no tool to list or inspect individual previously recorded claims/receipts, and no ability to retract an already-accepted claim, plus verification is limited to GitHub and npm among external registries.
Maintenance
Related MCP Connectors
Checks AI-written code for leaked keys and made-up packages, on your machine. Never uploads code.
Research-backed lint + generation for agent context files (CLAUDE.md, AGENTS.md) + prompt enhancer.
Scores your code across 12 areas, with findings, fix prompts, and re-assessment in your AI agent
Website checker for AI coding agents. Measures your site in a real browser, with evidence.
Related MCP Servers
- AlicenseBqualityDmaintenanceEnables auditing and verification of algorithmic trading backtests from coding agents like Claude Code, Cursor, and Windsurf, including look-ahead bias detection, overfitting checks, and sealed audit proof verification.3MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI coding agents to enforce spec-driven development and verify code before it is marked done, using six tools that catch invented APIs, scan for hallucinated content, check plugin conformance, sandbox-run tests, validate schemas, and record audit evidence.771 npm8PolyForm Noncommercial 1.0.0
- AlicenseNot gradedqualityCmaintenanceEnables auditing and drafting of AI agent files (CLAUDE.md, AGENTS.md, Cursor rules) and turning project descriptions into full build specs, with local and API-based tools.36 npmMIT
- AlicenseBqualityBmaintenanceChecks AI agents' claims against primary evidence, with no model; observes only what you grant.8108 PyPIApache 2.0