chore: Defend against AI-slop#994
Conversation
Assisted-by: Sisyphus:claude-sonnet-4-6 opencode
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughSummary by CodeRabbit
WalkthroughAdds AGENTS.md as the central agents reference, two SKILL docs (git-workflow and adding-skills), and pointer files ( ChangesAgent Skills Framework
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Suggested reviewers
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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. Comment |
There was a problem hiding this comment.
🧹 Nitpick comments (1)
.agents/skills/git-workflow/SKILL.md (1)
18-32: ⚡ Quick winConsider listing allowed commit types explicitly.
The document mentions "Conventional Commits format" but doesn't specify which types are valid. Based on the PR title validation workflow at
.github/workflows/pr-title.yml, the allowed types are:fix,feat,docs,ci,chore. Explicitly listing these in the guidance would help agents avoid using invalid types.📝 Suggested addition
## Commits -Conventional Commits format. Subject ≤50 chars. Body only when "why" isn't obvious. +Conventional Commits format. Subject ≤50 chars. Body only when "why" isn't obvious. + +Allowed types: `fix`, `feat`, `docs`, `ci`, `chore`.🤖 Prompt for 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. In @.agents/skills/git-workflow/SKILL.md around lines 18 - 32, Update the "Commits" section in SKILL.md to explicitly list the allowed Conventional Commit types used by our PR title validation workflow (allowed types: fix, feat, docs, ci, chore); edit the existing guidance under the "Commits" heading to add a short bullet or parenthetical that states these allowed types and notes that commit subjects must use one of them, and ensure the example commit follows that list (e.g., use feat or fix in the example) so agents produce valid commit types for the PR title validation workflow.Source: MCP tools
🤖 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.
Nitpick comments:
In @.agents/skills/git-workflow/SKILL.md:
- Around line 18-32: Update the "Commits" section in SKILL.md to explicitly list
the allowed Conventional Commit types used by our PR title validation workflow
(allowed types: fix, feat, docs, ci, chore); edit the existing guidance under
the "Commits" heading to add a short bullet or parenthetical that states these
allowed types and notes that commit subjects must use one of them, and ensure
the example commit follows that list (e.g., use feat or fix in the example) so
agents produce valid commit types for the PR title validation workflow.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 1be1736e-dd10-4111-bdd8-9d73427f9a44
📒 Files selected for processing (5)
.agents/skills/adding-skills/SKILL.md.agents/skills/git-workflow/SKILL.md.claudeAGENTS.mdCLAUDE.md
Assisted-by: Sisyphus:claude-sonnet-4-6 opencode
Assisted-by: Sisyphus:claude-sonnet-4-6 opencode
Assisted-by: Sisyphus:claude-sonnet-4-6 opencode
There was a problem hiding this comment.
Pull request overview
Adds project-specific agent/AI contributor guidance for pre-commit-terraform, including an AGENTS.md overview, a contributor-facing AI/LLM policy, and two agent skills describing workflow expectations.
Changes:
- Add
AGENTS.mdwith repo structure notes and hook/CLI contribution guidance. - Add
.github/AI_POLICY.mdand link it from.github/CONTRIBUTING.md. - Add
.agents/skills/documentation forgit-workflowandadding-skills.
Reviewed changes
Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| AGENTS.md | Adds agent-facing repo overview and contribution guidance (needs a couple doc accuracy fixes). |
| .github/CONTRIBUTING.md | Links contributors to the new AI policy. |
| .github/AI_POLICY.md | Introduces contributor-facing AI/LLM usage and quality expectations. |
| .agents/skills/git-workflow/SKILL.md | Documents commit/issue/PR workflow expectations including attribution rules. |
| .agents/skills/adding-skills/SKILL.md | Documents how to add/register skills and when to prefer scripts vs prose. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
|
|
||
| - **Model**: stated in system prompt as "You are powered by the model named X" — use X. Never hardcode. | ||
| - **Agent**: identity declared in system prompt (e.g. "Your designated identity is Sisyphus") — use that. Never use "Claude". | ||
| - **CLI**: check env vars — `OPENCODE=1` → `opencode`; `CLAUDE_CODE_SSE_PORT` set → `claude`; neither → omit CLI part. |
There was a problem hiding this comment.
| - **CLI**: check env vars — `OPENCODE=1` → `opencode`; `CLAUDE_CODE_SSE_PORT` set → `claude`; neither → omit CLI part. | |
| - **CLI**: check env vars — `OPENCODE=1` → `opencode`; `CLAUDECODE` set → `claude`; neither → omit CLI part. |
There was a problem hiding this comment.
It what was set in my setup, but it could be related to AWS Bedrock configuration.
Anyway, applied.
Co-authored-by: Anton Babenko <393243+antonbabenko@users.noreply.github.com>
|
This PR is included in version 1.107.0 🎉 |
Put an
xinto the box if that apply:Description of your changes
What
This section was generated by AI.
AGENTS.mdwith project-specific context (directory layout, hook patterns, repo rules, skill routing, AI policy pointer).agents/directory withcommands/,skills/,workflows/structureadding-skillsandgit-workflowskills with PR/commit/issue workflow rules.claude → .agentsandCLAUDE.md → AGENTS.mdsymlinksgit-workflowskill: AI attribution, WASTED guard for Why section, draft-only PR policy, attribution on all AI-written sections.github/AI_POLICY.md: ownership-first policy, disclosure encouraged, slop closed without discussionCONTRIBUTING.mdWhy
AI slop found us. Need to fight back. There are basically 2 levels of defences:
How can we test changes
This section was generated by AI.
Open the repo in an AI coding assistant —
AGENTS.mdand.agents/skills/are auto-loaded as context. Check.github/AI_POLICY.mdfor the contributor-facing policy.Assisted-by
Specific models used per commit are specified in the commit messages.