REA CI Tests Audit

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites — Reinforced framing — internal REA contributor workflow, read this first: this skill does not perform reverse engineering for you. It is a workflow for people who develop REA itself, for auditing REA's own CI and test suites (correctness, fixture drift, false passes, measured performance) inside a REA source checkout, following that checkout's AGENTS.md, testing guide and contribution rules. Installing this listing only copies the folder: no test is run, no CI is touched, and no configuration is written during installation. No account, key or payment is needed to copy the folder. Install "REA CI Tests Audit" for me. It is one of REA's internal contributor workflows — honest scope: it serves the people who build REA, not end users reverse engineering a target. It teaches an audit method for REA's own CI and test suites: establish the executed baseline at a pinned SHA (exact command, toolchain, host, selected tests), trace workflow triggers and test discovery to the commands that actually run, prove each suspected defect (fixture drift against current contracts, green tests that merely repeat their mock — checked by perturbing behavior in an isolated copy), and assess efficiency only on equivalent completed work, never by shrinking correctness coverage, suppressing failures, adding blanket retries or raising timeouts. The skill states that investigation alone does not authorize edits and grants no permission to publish issues, push or merge. Its folder holds the main SKILL.md, an OpenAI agent policy file, and a measurement-and-CI reference. Repository: https://github.com/morluto/rea (the skill is the whole .agents/skills/rea-ci-tests-audit/ folder at the pinned commit: SKILL.md, agents/openai.yaml and references/measurement-and-ci.md, 3 files in total) Pinned version: install the code at commit 6dc0f60faa0928e20b2395ba1a93ff8e497143f7 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/morluto/rea into a temporary folder, check out commit 6dc0f60faa0928e20b2395ba1a93ff8e497143f7, and copy the .agents/skills/rea-ci-tests-audit/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-22: [FINGERPRINTS] - SKILL.md: SHA-256 997f3717eacb261247b43805a75db9cbd6bd29eb88a5639b592b9705abac98ef - agents/openai.yaml: SHA-256 389cac55e7a6faf9b22e871b18adcdfd3bd89bdd00aa9e6636de07ca913ec916 - references/measurement-and-ci.md: SHA-256 691eb55373b24b8276dce5abcf94d9011b6ebeb7bdd9c23fb371654c5e52a583 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, and tell me in plain words: what the skill does when used, what it connects to, whether it reads or sends any personal information or credentials, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The folder holds 3 files: SKILL.md (6,786 bytes), agents/openai.yaml (display name and default prompt only), and references/measurement-and-ci.md on equivalent-work measurement and cache/artifact checks. There is no executable code in the folder. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. This install runs no test suite, fetches no CI results and modifies no REA checkout. - At use time the workflow runs inside a REA checkout I work on: it runs REA's own tests and CI checks locally under my existing authorization, keeps audit notes outside the tracked tree (normally /tmp), and separates fake-producer, boundary and actual-engine evidence. It reads no personal information or credentials of mine, and it grants no permission to publish issues, push or merge. - Do not run any test, audit or REA command, and do not use the skill in any way during this install: using it is a separate step that only starts when I ask for it later, inside a REA checkout. 3. When I say "yes, install", install it as a skill: copy the whole rea-ci-tests-audit folder into the agent's skills directory, in a folder named exactly "rea-ci-tests-audit", keeping its 3 files in their original relative layout (SKILL.md at its root). 4. Verify with no network calls: frontmatter valid (name: rea-ci-tests-audit), the 3 files in place with matching fingerprints. Do not run anything from the folder and do not use the skill to verify. 5. Report what was installed, where, and what I still need to do myself (for example: that this is an internal REA contributor workflow for auditing REA's own CI and test suites, that it is meant to be used inside a REA source checkout, and that using it is a separate step that only starts when I ask for it later). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.