I have an issue with what agets affirm, do you have too?

Elliot 0.3, an open-source MCP tool from developer EmanueleCell4, ran unattended inside Claude Code on 9 popular open-source AI coding-agent tools in October 2026 and produced 13 findings, of which an independent reviewer confirmed 12 correct and 0 wrong, at 1 to 3 minutes per repository. Elliot checks what an AI agent claims and what a README promises against the actual code, requiring every quote to exist at the cited line and commit, and it caught a tool whose README promised "No Telemetry" while shipping default-on analytics with consent pre-set to "yes" and another that re-enabled telemetry at login and identified users by email. On the same 9 repos Elliot found 6 of the 10 serious issues the author had found by hand, missing most often a missing authorization check, which the tool does not yet look for.

Independent verification for code written by AI agents. Elliot checks what your agent claims, and what your README promises, against the code. Every finding comes with proof. Quick start quick-start · What it found what-it-found · How it works how-it-works · Commands commands · Configuration configuration · FAQ faq In the repository you want to check: npx -y github:EmanueleCell4/elliot-mcp init Restart your agent, then type: /elliot-audit A 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. 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. 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. 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 | Wrong | Time per repo | |---|---|---|---| | 13 | 12 | 0 | 1 to 3 minutes | 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 https://github.com/winfunc/opcode/issues/481 - An app where logging in turns telemetry back on after the user switched it off, then identifies the user by email. public issue https://github.com/BloopAI/vibe-kanban/issues/3482 - 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 https://github.com/dyad-sh/dyad/issues/4879 , PR https://github.com/dyad-sh/dyad/pull/4881 - 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. 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 roadmap . 1. 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. 2. Read and propose your agent . Your agent traces each claim into the code and proposes a verdict, quoting the docs and the code. 3. 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. 4. 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. 5. 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. 6. 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. The full report is in examples/lex-ai-report.html https://github.com/EmanueleCell4/elliot-mcp/blob/main/examples/lex-ai-report.html . It was produced by an earlier version, before the independent review step existed. 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 three init 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 | .mcp.json project | .claude/commands/elliot-audit.md → /elliot-audit | | Cursor | .cursor/mcp.json project | .cursor/commands/elliot-audit.md → /elliot-audit | | Codex | ~/.codex/config.toml | ~/.codex/prompts/elliot-audit.md → /prompts: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 . 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" | elliot snapshot , then verify changes | the content hash of every file before and after the task | | "It's deployed / live" | verify url | the HTTP status and the text really served at the URL | | "PR merged", "CI is green", "pushed" | verify github | the GitHub API: merge state, check runs, whether the commit exists | | "Published to npm" | verify npm | the npm registry: version exists, published after a given time | | "What did my agent really do today?" | elliot report | 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. | Tool | Used by | What it does | |---|---|---| | elliot audit start | the agent | Pins the commit, lists the files that make claims, runs the risk sweep, returns the protocol. | | elliot record claim | the agent | Records one claim with a verdict. Checks the source quote, every evidence quote and every absence search. | | elliot record behavior | the agent | Records risky code behaviour the docs never mention, with checked code quotes. | | elliot review packet | the agent | Returns a self-contained prompt for a fresh reviewer subagent. | | elliot record review | the reviewer | Confirms, downgrades or retracts a finding. A retraction needs a code quote. | | elliot audit finish | the agent | Refuses to finish while risk-sweep items or reviews are missing, then writes the report. | | elliot snapshot | the agent | Fingerprints the working tree before a task. | | verify changes | the agent | Compares the tree with a snapshot, or with HEAD. | | verify url | the agent | Checks status and expected text at a URL. | | verify github | the agent | Checks pr merged , checks passing or commit exists . | | verify npm | the agent | Checks a package version on the registry. | | elliot report | you | Summarises recent receipts. | | Verdict | Meaning | |---|---| | VERIFIED | The code implements the claim. A small gap that misleads nobody is kept as a note. | | PARTIAL | Implemented, with a gap that would make a user who relies on it do something wrong. | | FALSE | The code contradicts the claim, or the feature, command, flag or option does not exist. | | UNVERIFIABLE | Depends on a deployment, a contract or a service outside the repo. | | UNDISCLOSED | Risky behaviour in the code that the docs never mention. | Use this if you'd rather not run init . Claude Code claude mcp add elliot -- npx -y github:EmanueleCell4/elliot-mcp Codex ~/.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 https://github.com/EmanueleCell4/elliot-mcp/blob/main/skills/elliot/SKILL.md . All settings are optional environment variables. | Variable | Default | What it does | |---|---|---| | GITHUB TOKEN | unset | Used by verify github for private repos and higher rate limits. | | ELLIOT DIR | .elliot/ in the working directory | Where receipts and audit state are stored. | | ELLIOT GITHUB API | https://api.github.com | GitHub API base URL, for GitHub Enterprise or tests. | | ELLIOT NPM REGISTRY | https://registry.npmjs.org | Registry used by verify npm . | 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 state Add .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 makes a receipt tamper-evident: change a field and the digest no longer matches. A local receipt is still produced on your own machine, so it proves nothing to a third party. Independently signed receipts are on the roadmap. 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. Elliot is a static reading of the code. It does not run your app yet, and it does not look for missing authorization checks 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. - File, URL, GitHub and npm checks with local receipts - /elliot-audit : claim audit with checked citations, report in elliot-report/ - Risk sweep, undisclosed behaviours, materiality test and independent review - Authorization checks: unauthenticated handlers, missing ownership checks - Verify the agent's own "done" at the end of every task, by running tests and the app - Hooks that run Elliot automatically when the agent stops Claude Code, Codex - A GitHub Action that audits the claims touched by each pull request - Independently signed receipts with a public verification link git clone https://github.com/EmanueleCell4/elliot-mcp && cd elliot-mcp npm install npm test 19 end-to-end tests: spawns the server over stdio and drives it with the MCP client The server is plain Node ES modules with two dependencies: the MCP SDK and zod. The main pieces: | File | What it does | |---|---| | src/index.js | MCP server, tool definitions, init entry point | | src/audit.js | Audit state, citation checks, review gate, verdict resolution | | src/risk.js | Risk sweep: risky doc promises and code signals | | src/audit-protocol.js | The protocol the agent follows | | src/audit-report.js | HTML and Markdown reports | | src/init.js | Registers Elliot in Claude Code, Cursor and Codex | | src/checks/ | verify changes , verify url , verify github , verify npm | Issues, false positives and stories of agents telling you they were done are very welcome. MIT https://github.com/EmanueleCell4/elliot-mcp/blob/main/LICENSE © 2026 Emanuele Cella