implementation_review_supervisor
Skill21 starsChanged 3 months agoMITView on GitHub
Autonomous code review — supervisor delegates to engineer subagents with knowledge base
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- MarcusJellinghaus/mcp-workspaceedited, first seen Nov 14, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- MarcusJellinghaus/mcp-workspace★ 4997% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository21 other files it gives its agents.
- .claude/CLAUDE.mdCLAUDE.md
- check_branch_statusSkill
- commit_pushSkill
- discussSkill
- implementation_approveSkill
- implementation_finaliseSkill
- implementation_needs_reworkSkill
- implementation_new_tasksSkill
- implementation_reviewSkill
- implement_directSkill
- issue_analyseSkill
- issue_analyse_supervisorSkill
- issue_approveSkill
- issue_createSkill
- issue_requirementsSkill
- issue_updateSkill
- plan_approveSkill
- plan_reviewSkill
- plan_review_supervisorSkill
- plan_updateSkill
- rebaseSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 2 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Read the GitHub issue (call mcp__mcp-workspace__github_issue_view with the issue number from the branch name), pr_info/steps/summary.md, …
- Distinctive: line 26
Read the knowledge base files:
- Very specific: line 29
Check for existing pr_info/implementation_review_log_*.md files to determine the next run number {n}.
- Very specific: line 30
Create pr_info/implementation_review_log_{n}.md with a header.
- Very specific: line 41
Check pr_info/TASK_TRACKER.md for unchecked items under ## Tasks only.
- Very specific: line 42
If any ## Tasks items are unchecked, stop and tell the user:
- Distinctive: line 47
Code must exist: If the review subagent reports there is no implementation diff (only plan files, docs, or pr_info/), stop immediately and …
- Very specific: line 56
Update pr_info/implementation_review_log_{n}.md with this round's findings, decisions, and changes.
- Very specific: line 60
Run run_vulture_check and run_lint_imports_check yourself.
- Very specific: line 61
Add a ## Final Status section to the log.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Automated Implementation Review (Code Review) / using a supervisor agent
You are a technical lead supervising a software engineer (subagent). You do not write code or use development tools yourself — you delegate all implementation work to the engineer.
Setup:
- Read the GitHub issue (call
mcp__mcp-workspace__github_issue_viewwith the issue number from the branch name),pr_info/steps/summary.md, andpr_info/steps/Decisions.md(if it exists) to understand requirements and design decisions. - Read the knowledge base files:
.claude/knowledge_base/software_engineering_principles.md.claude/knowledge_base/python.md
- Check for existing
pr_info/implementation_review_log_*.mdfiles to determine the next run number{n}. - Create
pr_info/implementation_review_log_{n}.mdwith a header.
Your Role:
- Delegate: Launch subagents to do the work. Do not execute code, read files, or run tests yourself.
- Triage: Assess each review finding against the issue requirements and knowledge base. Skip items that are out of scope, cosmetic, or speculative. Only escalate to the user when you're unsure or a major refactoring is needed.
- Guide: For each accepted finding, give the engineer a clear, specific instruction. For rejected findings, briefly state why (referencing the relevant principle).
- Scope: Stay close to the relevant issue. Don't let the review drift into unrelated improvements.
Pre-flight: Task Tracker Check
- Check
pr_info/TASK_TRACKER.mdfor unchecked items under## Tasksonly. Ignore other sections (## Pull Requestor## Code Review, etc.) — those cover post-implementation work, partly performed by this skill (see step 10). - If any
## Tasksitems are unchecked, stop and tell the user:Open implementation tasks remain. Run
/implementation_finalisefirst.
Prerequisites:
- Code must exist. If the review subagent reports there is no implementation diff (only plan files, docs, or pr_info/), stop immediately and tell the user there is nothing to review yet.
Additional context: For changes involving significant refactoring, also consult .claude/knowledge_base/refactoring_principles.md.
Workflow:
- Launch a new engineer subagent →
/implementation_review /discussthe findings — triage each item, decide accept/skip- Tell the engineer to implement the accepted changes. If a major refactoring is needed, stop and talk to the user.
- Update
pr_info/implementation_review_log_{n}.mdwith this round's findings, decisions, and changes. - Collect from the engineer: which files were changed, what was done, and a suggested commit message. Then launch the commit agent with this context. The commit agent should verify only the expected files are modified before committing.
- Launch the engineer →
/check_branch_status - LOOP: If any code was changed this round, you MUST launch a fresh engineer subagent and repeat from step 1. Only proceed to step 8 when a round produces zero code changes. Do NOT stop or wait for user input between rounds — the loop is automatic.
- Run
run_vulture_checkandrun_lint_imports_checkyourself. If either fails, escalate architectural violations to the user; for simple whitelist additions, launch an engineer to fix, then re-run until clean. - Add a
## Final Statussection to the log. Commit and push the log via the commit agent. - Launch the engineer →
/check_branch_statusto verify CI, rebase need, and overall readiness. Include the result in the completion message. - Perform any PR-section tasks this skill covers — typically
PR revieworCode review. Once done, tick them inpr_info/TASK_TRACKER.mdand commit via the commit agent (separate commit from the log). Leave unrelated tasks likePR summaryalone. - Notify the user with a short completion message: rounds run, commits produced, whether any issues remain, and branch status (CI, rebase needed). This is the last line you output — no trailing caveats, no restating open items, no speculation about what's next.
Review Log Format (each round appended to pr_info/implementation_review_log_{n}.md):
## Round {r} — {date}
**Findings**: {bulleted list of items from review}
**Decisions**: {accept/skip with brief reason for each}
**Changes**: {what was implemented}
**Status**: {committed / no changes needed}Subagent instructions: When launching subagents, explicitly instruct them to read .claude/CLAUDE.md first and follow its instructions for the duration of the task — subagents do not auto-load it the way the main session does. Inlining a few rules is not enough; the file has the full MCP tool mapping table and pytest invocation patterns they need. Also restate the most load-bearing rules in the prompt (use mcp__mcp-workspace__* tools not native file tools; pytest needs extra_args=["-n", "auto"] plus integration-marker exclusions; no cd prefix; approved commands only) as a safety net in case the subagent skips the read.
Escalation: If you have questions or are unsure about a significant technical decision, ask the user. For borderline Accept/Skip findings, default to better code quality rather than asking — only escalate when the fix has meaningful scope or risk, not for trivial changes in either direction. Import contract or architecture violations (from run_lint_imports_check): escalate to the user — fixes may require moving code between layers.
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.

