Codex Autoresearch is an autonomous experimentation system for Codex that repeatedly changes a repository, measures the result, and keeps or reverts each change. It is for optimizing measurable outcomes such as test failures, coverage, type errors, warnings, latency, binary size, or reproducible security findings. The catalogue entry provides the Codex skill that runs and manages these experiment loops.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add leo-lilinxiao/codex-autoresearch --skill codex-autoresearchgit clone --depth 1 https://github.com/leo-lilinxiao/codex-autoresearchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/leo-lilinxiao/codex-autoresearch/codex-autoresearch)<a href="https://agentmods.dev/skills/leo-lilinxiao/codex-autoresearch/codex-autoresearch"><img src="https://agentmods.dev/badge/skills/leo-lilinxiao/codex-autoresearch/codex-autoresearch/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/leo-lilinxiao/codex-autoresearch/codex-autoresearch"><img src="https://agentmods.dev/badge/skills/leo-lilinxiao/codex-autoresearch/codex-autoresearch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00080 | $0.01622 |
| Opus 5 | $0.00040 | $0.00811 |
| Sonnet 5 | $0.00016 | $0.00324 |
| Haiku 4.5 | $0.00008 | $0.00162 |
Grade A, and why
codex-autoresearch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Autoresearch
Turn a repo-level goal into a controlled loop:
inspect -> change one thing -> verify -> keep or revert -> repeat
Codex supplies the engineering judgment. The bundled control script supplies strict Git boundaries, measurement, rollback, state, and logs.
Load
- Read
references/workflow.mdfor every invocation, including status, history, report, stop, and resume. - Read
references/experiment.mdbefore starting or continuing an active run. - Read
references/background.mdonly for a background run.
Resolve commands from this skill's own directory as <skill-root>/scripts/autoresearch.py. Never assume the target repository contains the script.
Before Starting
-
Require one Git repository root. If the task spans repositories, ask the user to choose one run per repository.
-
Check for
autoresearch-results/run.jsonwith:python3 <skill-root>/scripts/autoresearch.py status --repo <repo>not_initializedis fresh. Any other status or schema error must be surfaced; do not infer state from other files. -
For a fresh run, inspect the repo and propose:
- one plain-language goal,
- repository-relative file or directory scopes (no globs),
- one numeric metric and whether lower or higher is better,
- a command whose final non-empty stdout line is that number, or a JSON object plus one explicit key,
- a numeric target,
- an optional baseline-passing guard command,
- foreground or background,
- an optional iteration limit.
-
Run candidate measurement commands read-only if needed, then show one concise confirmation. Include the baseline, target, scope, commands, mode, and the fact that each trial is committed and failed trials are reverted.
-
Do not write project files, initialize artifacts, create a Goal, or launch a controller before clear user approval such as
go.
Start
After approval, use the exact confirmed values.
Foreground
Initialize once:
python3 <skill-root>/scripts/autoresearch.py init \
--repo <repo> --goal <goal> --scope <path> \
--metric-name <name> --direction <lower|higher> \
--verify <command> [--metric-key <key>] --target <number> \
[--guard <command>] [--max-iterations <n>]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 132 lines · 80 tokens per session scan A 9bf0d6936d50
codex-autoresearch is a skill published in the GitHub repository leo-lilinxiao/codex-autoresearch (2,475 stars, last pushed 1mo ago), licensed MIT. It adds 80 tokens to every session and 1,622 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cap-commit-push
Use when the user sends cap or asks for a fast commit-and-push shortcut for current repository changes.
versioning-by-impact
Use when completing a task and deciding whether repository changes require a semantic version bump and changelog entry before commit/push.
codexkit-repository-maintenance
Use when maintaining or improving the GameStudio-CodexKIT source repository, including CI, governance, catalog, generators, adapters, packaging, documentation, versioning, or release readiness.
studio-workspace-routing
Use when .agents/project-profile.yaml, nested Git roots, or cross-repository game work requires a profile-defined repository route, validation slice, or cross-project phase.
release-candidate-preflight
Use when a release candidate or RC needs a go/no-go readiness decision across required artifacts, test summaries, known issues, approvals, compatibility, monitoring, and rollback.
scenario-readiness-review
Assess whether a scenario's uncommitted changes are ready to commit by querying the git-control-tower unified review endpoint for code health signals, then reviewing the actual git changes for coherence and intent.