review-loop plugin
Code review for teams where agents write most of the code: route changes by risk, review against the spec and the evidence, and turn reviews into teaching. All three share a REVIEW-LESSONS.md memory that people teach and agents read.
How to install the review-loop plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add R-Hatim/hrih-skills
/plugin install review-loop@hrih-skillsWhat's inside review-loop
3 parts read from the plugin's own files.
Skills 3
Claude loads it when the task matches- review-apprentice
Turn code review into teaching when an AI reviews alongside people, so juniors learn to review instead of rubber-stamping the agent. Use it when a junior or new team member reviews a pull request with an AI reviewer, when a senior wants to mentor through reviews, or for requests like 'review this PR with me', 'help me learn to review', 'check my review', 'what did I miss?', 'train me on this codebase', and when a solo developer wants to keep their own reviewing sharp. The junior reviews first without seeing the AI's findings, then accepts or rejects each AI finding with a reason, then a senior (or, solo, the lessons file) checks those calls. Every correction becomes a plain-language lesson in REVIEW-LESSONS.md that the next reviewer, human or agent, reads. Even when asked to just give the answers, it offers hints first and the list on request. Skip it when the goal is fast triage of many changes (use review-router) or nobody wants to learn from the review.
- review-router
Decide how much human review a code change needs, judged from the change itself rather than its author, and tell the human exactly what to read first. Use it to triage, route or risk-rate a pull request, merge request, branch, commit range or diff, and for questions like 'does this need a human?', 'how risky is this PR?', 'can this auto-merge?', 'who should look at this?', or when agent-written changes arrive faster than people can read them. It sets a tier (low, medium or high) from a deterministic floor of sensitive paths plus judgement about intent, blast radius and reversibility, pairs every risk with the check that covers it, points at the three hunks to read first, and learns from human overrides through a REVIEW-LESSONS.md file read from the base branch. Skip it for a line-by-line bug hunt (use a code review tool), for style or lint questions, and for code that isn't in git.
- spec-review
Review a code change by its spec and its evidence instead of reading every line: check it does what was asked, prove it with tests that ran, and find what the tests don't pin down. Use it when an agent wrote the code and a human wants to know whether it does what the ticket, issue, spec.md, requirements or acceptance criteria say, and for requests like 'does this PR do what the ticket says?', 'review against the spec', 'are these tests any good?', 'what didn't the agent test?', 'check the acceptance criteria'. It restates the spec as numbered testable criteria for the human to confirm, writes acceptance checks from the spec without looking at the code, maps every criterion to tests and real output, runs a small mutation pass over the changed lines, and flags tests bent to fit the code. Skip it when there's no spec, ticket or stated intent to check against (use a normal code review), and for pure refactors that shouldn't change behaviour.
Using review-loop in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update review-loop@hrih-skills
- Refresh the marketplace
- /plugin marketplace update hrih-skills
- Turn it off for now
- /plugin disable review-loop@hrih-skills
- Remove it
- /plugin uninstall review-loop@hrih-skills
Skills in review-loop
- review-apprenticeR-Hatim/hrih-skills
Turn code review into teaching when an AI reviews alongside people, so juniors learn to review instead of rubber-stamping the agent. Use it when a junior or new team member reviews a pull request with an AI reviewer, when a senior wants to mentor through reviews, or for requests like 'review this PR with me', 'help me learn to review', 'check my review', 'what did I miss?', 'train me on this codebase', and when a solo developer wants to keep their own reviewing sharp. The junior reviews first without seeing the AI's findings, then accepts or rejects each AI finding with a reason, then a senior (or, solo, the lessons file) checks those calls. Every correction becomes a plain-language lesson in REVIEW-LESSONS.md that the next reviewer, human or agent, reads. Even when asked to just give the answers, it offers hints first and the list on request. Skip it when the goal is fast triage of many changes (use review-router) or nobody wants to learn from the review.
5 files6stars - review-routerR-Hatim/hrih-skills
Decide how much human review a code change needs, judged from the change itself rather than its author, and tell the human exactly what to read first. Use it to triage, route or risk-rate a pull request, merge request, branch, commit range or diff, and for questions like 'does this need a human?', 'how risky is this PR?', 'can this auto-merge?', 'who should look at this?', or when agent-written changes arrive faster than people can read them. It sets a tier (low, medium or high) from a deterministic floor of sensitive paths plus judgement about intent, blast radius and reversibility, pairs every risk with the check that covers it, points at the three hunks to read first, and learns from human overrides through a REVIEW-LESSONS.md file read from the base branch. Skip it for a line-by-line bug hunt (use a code review tool), for style or lint questions, and for code that isn't in git.
5 files6stars - spec-reviewR-Hatim/hrih-skills
Review a code change by its spec and its evidence instead of reading every line: check it does what was asked, prove it with tests that ran, and find what the tests don't pin down. Use it when an agent wrote the code and a human wants to know whether it does what the ticket, issue, spec.md, requirements or acceptance criteria say, and for requests like 'does this PR do what the ticket says?', 'review against the spec', 'are these tests any good?', 'what didn't the agent test?', 'check the acceptance criteria'. It restates the spec as numbered testable criteria for the human to confirm, writes acceptance checks from the spec without looking at the code, maps every criterion to tests and real output, runs a small mutation pass over the changed lines, and flags tests bent to fit the code. Skip it when there's no spec, ticket or stated intent to check against (use a normal code review), and for pure refactors that shouldn't change behaviour.
6 files6stars
review-loop plugin questions
How do I install the review-loop plugin?
In Claude Code run /plugin marketplace add R-Hatim/hrih-skills, then /plugin install review-loop@hrih-skills.
What does the review-loop plugin include?
review-loop bundles 3 skills.
How do I update or remove the review-loop plugin?
Update it with claude plugin update review-loop@hrih-skills in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall review-loop@hrih-skills.