parler-review
Skill15 starsChanged 4 months agoApache-2.0View on GitHub
Review a diff, branch, or PR in the Parler Protocol repo against the project review contract (docs/code-review-guidelines.md). Use when asked to review changes, audit a branch, check a PR, or self-review before landing. Produces verified, severity-ranked findings — no style noise, no unverified criticals.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- architect-researchSkill
- architectSkill
- code-standardsSkill
- validate-marketSkill
- write-blogSkill
The file
Files
Review a Parler Protocol change
Execute docs/code-review-guidelines.md — read it now
if it isn't in context. That document is the contract; this skill is the runbook.
- Scope:
git diff origin/main...plusgit status(include working-tree changes). For a PR:gh pr diff <n>. - Load: skim "Hard gates" and "Invariants" in
docs/engineering-guidelines.md; checktasks/lessons.mdfor traps in the touched files. - Read every changed function in the file, full context — never review from diff hunks alone.
- Verify every candidate finding: production-reachable (not
#[cfg(test)]), concrete failure scenario, cited lines actually read. Drop or downgrade what you can't prove. - Run the gates for non-doc changes:
scripts/verify.sh, ormake ciwhen more than one crate is touched. - Report: severity-ranked findings in the contract's format, then a verdict (approve / approve with nits / needs changes) plus gate results.
This is a read-only pass: never modify files during the review; Bash is for git, gh, and the
gates. For a large diff, offload to the code-reviewer agent (.claude/agents/code-reviewer.md)
to keep the main context clean.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

