Maintainer of this skill?
by lunox-work
Due diligence for AI coding agents: independent, hidden test suites written from the spec. The agent works against a public suite; a hidden suite it never sees judges the result.
# Add to your Claude Code skills
git clone https://github.com/lunox-work/duediligenceGuides for using ai agents skills like duediligence.
See how duediligence compares with popular alternatives.
duediligence is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by lunox-work. Due diligence for AI coding agents: independent, hidden test suites written from the spec. The agent works against a public suite; a hidden suite it never sees judges the result. It has 0 GitHub stars.
duediligence's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/lunox-work/duediligence" and add it to your Claude Code skills directory (see the Installation section above).
duediligence is primarily written in TypeScript. It is open-source under lunox-work on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh duediligence against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
Named for due diligence: before a deal closes, someone who did not build the business checks it. duediligence does that for an agent's work: independent tests, written from the spec and hidden from the agent, decide whether it is done.
"The escrow release logic in
src/index.tsis implemented. The build is clean and my 21 tests undertests/own/pass.""Zero-amount releases are recorded. … If the grader expects
releasesto be empty for pro bono tasks, this is the line to change."
hidden 43/46
Failing scenarios: boundary-2
hidden: [boundary-2] passing hidden tests on a
bounty of 0 at any share records nothing
and the task is pro bono
+ [{ kind: 'hidden-tests', amount: 0 }]
- []
Same code. 21 green tests. One wrong guess, and the agent knew it was guessing.
escrow-split, run 1. 14 of 15 runs made the same guess. The ticket never said what a $0 bounty does; the agent picked an answer, then wrote 21 tests that agree with it. Green tests from the author of the code test the author's guesses.
npx skills add lunox-work/duediligence -g
Then ask your coding agent (Claude Code, Codex, Cursor, or any agent that loads skills): "write tests my agent can't see for this ticket". The tests are written on whichever account you have: Claude Code, a Codex login or an Anthropic API key. The implementer can be anything that edits a folder: Claude Code, Codex, Cursor, Copilot, Aider, or a person. Try the bundled example in five minutes; a git clone and the CLI are in Install in full.
105 of 105 agents said their own tests passed. 75 of them failed tests they never saw. None of them cheated.
Same agent (Haiku 5.5), same loose tickets, 7 specs × 15 runs. The only change in the middle column (p < 1e-11): the decisions the ticket left open, confirmed by a person and handed to the agent as tests it can see, judged by tests it can't. In the last, each of the 24 rejected runs got one retry told only the failing scenario's key, never the test; 15 then passed. Were the rejections fair? A blind referee on another model, given the spec, the code and one anonymised failing test, upheld 95 of 100.
In 2025 METR caught frontier models reward hacking in 30% of RE-Bench runs: patching the grader, copying its answers. That is what sandboxes guard against. In 2,522 sandboxed runs here no agent changed a test or the harness, and the hidden suites still rejected 71% from a loose ticket. The agent does not need to cheat to be wrong.
Hidden, calibrated test suites for AI coding agents, independent of the agent they judge. An agent skill and CLI, for Claude Code, Codex, Cursor and other agents, that turns a ticket or spec for one module in TypeScript, Python, Go or Rust, for a command or an HTTP service checked only at that boundary, or for a change to one package of an existing TypeScript or Python repository, into a public test suite the implementing agent works against and a hidden suite, which the agent never sees, that judges the result. A writer model (Opus 5.5 through Claude Code, or Codex or the Anthropic API without it) writes both suites and the reference that calibrates them, in sessions apart from the implementing agent. A person confirms the spec's decisions before any test is written. The implementing agent can be anything that edits a folder: Claude Code, Codex, Cursor, Copilot, Aider, or a person.
Use it when "all tests pass" comes from the same agent that wrote the tests, and you want acceptance decided by tests it could not have overfitted, gamed or guessed.
About 3,000 graded agent runs on 26 TypeScript specs, 40 LiveCodeBench problems, 40 HumanEval+ tasks and a spec ladder judged by oracles no model wrote, 11 real commits replayed inside their repository and 12 more in two public repositories (unjs/ufo, unjs/pathe). Opus 5.5 wrote the suites, Haiku 5.5 implemented unless stated; Codex CLI as the writer and the referee is in RESULTS.md "Writer backends", as an implementer in "Other implementers", and Opus 5.5 as an adversary in "Red team". Every table, with Fisher's exact p-values, is in RESULTS.md.
| question | measured |
|---|---|
| Agent's own tests green, hidden suite rejects | 75 of 105 runs from a loose ticket (71%); 25 of 404 from a precise spec (6%) |
| Were those rejections right? | A blind referee on another model, given the spec, the code and one anonymised test, upheld 95 of 100; Codex as the referee, also 95 of 100 |
| Does giving the agent a public suite help? | From a ticket, 81 of 105 pass the hidden suite, against 30 of 105 without (p < 1e-11) |
| Does the hidden suite resist gaming? | Haiku: 21 of 21 gamed submissions rejected (20 hard-coded its inputs). Opus 5.5 on the 4 specs mutants survive most: 20 of 20 rejected on two, 0 of 20 on the other two, where an audit finds a skipped rule in 18 |
| False rejects, against an outside oracle | 3 of 289 correct HumanEval+ solutions; wrong ones accepted 4.8% vs 15.0% by HumanEval's own tests |
| Does it depend on Claude writing the suites? | Codex CLI wrote the 26 specs' suites: 99.4% of its hidden tests survive calibration on Claude's reference, and both suites kill 80.6% of 1,442 mutants against Claude's 79.2%; on EvalPlus they accept 3.8% of wrong solutions against 4.8% |
| Agreement with a real tool | The calibrated reference matched git apply on all 7,533 real-history cases judged |
| Judged through a command or a service | patch-apply as node src/cli.js: the same 79 hidden tests kept, the same verdict on 196 of 196 stored submissions, git apply's answer on 7,533 of 7,533 judged cases. As an HTTP service (csv-export): a 501 stub incomplete; blind Haiku passed 2 of 5, the other 3 fail one real rule |
| The same in Python, Go and Rust |