Simplify Monday agent guidance and add read-only audit Skills

[proerror77]

Change contract

Replace the verbose root agent manual with a concise evidence-and-authority router, and package the three recurring read-only audit procedures as project Skills.

Out of scope

Runtime cutover, deployment mutation, scheduled automation, issue-lifecycle replacement, branch/worktree cleanup, trading code, and research code.

Dependency or merge order

None. This is one agent-guidance rollout and can be reverted as one documentation commit set.

Focused validation

• Official quick_validate.py passed for all three Skills.
• Root instruction content contract and every referenced repository path passed.
• Managed-worktree preflight passed.
• Worktree Skill smoke report completed: 65 active, 4 dirty, 0 Git-prunable.
• git diff --check passed.
• Standards review: no findings.
• Spec review against Monday: concise agent router and delivery-status Skill #640, Monday: research evidence audit Skill #641, and Monday: worktree audit Skill #642: no findings.

Rollout/rollback impact

Future sessions load the concise root instructions and discover the project Skills after merge. Existing sessions must be told to reread the landed file. Rollback is a normal revert; no runtime state changes.

Issue relationship

Refs #639

Summary by CodeRabbit

• New Features

  • Added guided, read-only workflows for verifying delivery status, research evidence, and worktree health.
  • Workflows provide structured results, identify incomplete checks, and stop when evidence is unavailable or changes are unsafe.

• Documentation

  • Added concise operating guidance covering workflow selection, evidence validation, rollout controls, ownership checks, and focused validation.
  • Added descriptions and default prompts for the new workflow tools.

[coderabbitai]

Warning

Review limit reached

@proerror77, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 46 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 9f190bc8-f643-4298-95e8-bf6235ad83f3

📥 Commits

Reviewing files that changed from the base of the PR and between 1a3a82a and 00c6220.

📒 Files selected for processing (3)

• .agents/skills/monday-delivery-status/SKILL.md
• .agents/skills/monday-research-evidence-audit/SKILL.md
• .agents/skills/monday-worktree-audit/SKILL.md

📝 Walkthrough

Walkthrough

Replaced AGENTS.md with concise Monday operating rules. Added three read-only skills for delivery status, research evidence, and worktree audits. Added OpenAI metadata for each skill.

Changes

Monday operating guidance

Layer / File(s)	Summary
Root operating guide AGENTS.md	Defines authority boundaries, workflow routing, evidence and readback rules, worktree and branch safeguards, validation guidance, and skill-extraction rules.
Delivery status audit .agents/skills/monday-delivery-status/SKILL.md, .agents/skills/monday-delivery-status/agents/openai.yaml	Defines evidence checks for code, CI, merge, release, runtime, and readback states. Adds read-only controls and structured status output.
Research evidence audit .agents/skills/monday-research-evidence-audit/SKILL.md, .agents/skills/monday-research-evidence-audit/agents/openai.yaml	Defines authenticated evidence tracing through snapshots, Mission admission, evaluation, publication, and readback.
Worktree audit .agents/skills/monday-worktree-audit/SKILL.md, .agents/skills/monday-worktree-audit/agents/openai.yaml	Defines worktree inventory, classification, ownership checks, cleanup-safety checks, and destructive-operation stop conditions.

Estimated code review effort: 2 (Simple) | ~15 minutes

Possibly related issues

• PRD: Simplify agent guidance and extract recurring read-only audits #639: Covers the AGENTS.md simplification and extraction of the three read-only audit workflows.
• Monday: concise agent router and delivery-status Skill #640: Covers the concise root guidance and delivery-status skill.
• Monday: research evidence audit Skill #641: Covers the research evidence audit skill.
• Monday: worktree audit Skill #642: Covers the worktree audit skill.

Possibly related PRs

• proerror77/monday#79: Both changes modify AGENTS.md with agent operating guidance.
• proerror77/monday#342: Both changes define Git worktree, branch, and cleanup-safety rules.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main changes: simplifying Monday agent guidance and adding read-only audit skills.
Description check	✅ Passed	The description covers the change, scope, dependencies, validation, rollout, rollback, and issue relationship; only the optional scope exception section is missing.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches 💡 1

🛠️ Fix failing CI checks 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/simplify-agents-md

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In @.agents/skills/monday-delivery-status/SKILL.md:
- Around line 15-16: Update the Runtime and Readback contract in the deployment
verification instructions so Readback is correlated with the exact deployed
immutable identity recorded for Runtime, including image/SHA and configuration
identity. Require reporting Runtime or Readback as unknown when that identity
correlation is unavailable, and preserve the independent external
artifact/service readback requirement.

In @.agents/skills/monday-worktree-audit/SKILL.md:
- Around line 16-20: Update the State classification contract in the preflight
report, specifically the report() function: keep State mutually exclusive based
only on registration/cleanliness and Git’s prunable status, with dirty taking
precedence for any tracked or untracked changes and active limited to clean,
non-prunable worktrees. Record ownership or session use only in Owner/use, and
do not allow those conditions to alter State or produce conflicting
classifications.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 68da61b9-83d3-430d-8a4d-6d32c18d154d

📥 Commits

Reviewing files that changed from the base of the PR and between 48748c5 and 1a3a82a.

📒 Files selected for processing (7)

• .agents/skills/monday-delivery-status/SKILL.md
• .agents/skills/monday-delivery-status/agents/openai.yaml
• .agents/skills/monday-research-evidence-audit/SKILL.md
• .agents/skills/monday-research-evidence-audit/agents/openai.yaml
• .agents/skills/monday-worktree-audit/SKILL.md
• .agents/skills/monday-worktree-audit/agents/openai.yaml
• AGENTS.md

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 1a3a82a6ef

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".