Pm Review

PM Review

Orchestrator skill for PMBOK-based review of local git changes. Gathers the change set + project context, then hands off to the @agent-pm-review:pm-review subagent for the analysis itself.

Arguments

Workflow

1. Collect Changes ──► 2. Read Context ──► 3. Delegate to subagent ──► 4. Surface report

Phase 1: Collect changes

git diff --cached                  # staged (always)
git diff                           # unstaged (skip if --staged-only)
git status --porcelain | grep '^??' # untracked (skip if --staged-only)

If everything is empty: tell the user "No local changes detected. Nothing to review." and exit.

Feedback Check

If feedback/log.md exists and has 5+ entries, read the last 10. If a pattern is apparent (same issue in 3+ entries, or avg rating < 3): tell the user "Recurring feedback detected: [brief]. Consider running /skill-improve --skill pm-review." Continue with normal execution.

Phase 2: Read project context

For each changed file, read the full file. Also gather project-level docs that inform PM judgment:

• README.md, CHANGELOG.md
• Package manifests (package.json, pyproject.toml, etc.)
• CI/CD configuration files
• Any roadmap or requirements docs the repo carries

Phase 3: Delegate to the subagent

Hand the change set + the context bundle to @agent-pm-review:pm-review. Pass through the --staged-only and --file flags so the subagent can write the report file directly.

The subagent applies the PMBOK lens across the 7 knowledge areas (Scope, Risk, Stakeholder, Quality, Integration, Schedule, Resource) and returns a structured report.

Phase 4: Surface the report

Print the subagent's report inline. If --file was passed and the subagent already wrote the file, confirm the path. Otherwise, write the report to pm-review-YYYY-MM-DD-HHmm.md.

Retrospective

After completing the workflow, reflect on the entire execution session:

1. Were there mid-session corrections, rejected outputs, plan changes, or errors?
2. Ask the user: "Quick feedback on this run? (1-5 rating, note any issues, or press enter to skip)"
3. If the user provides feedback OR if corrections/issues occurred: a. Create feedback/ if it does not exist (header: # Feedback Log + blank line + <!-- Append new entries at the top. Do not edit previous entries. -->). b. Prepend a new entry:

   ## <ISO-8601 timestamp>
   - **Skill Version**: <version from this file's frontmatter>
   - **Task**: <brief description>
   - **Outcome**: success | partial-success | failure | error
   - **Rating**: <N>/5 (or "—" if not provided)
   - **Corrections**: <mid-session corrections, or "none">
   - **Issues**: <specific problems, or "none">
   - **User Note**: <user's verbatim feedback, or "—">
   ---
   

4. If the user skips AND no corrections / issues occurred, end without recording.

Behavior Scenarios

Scenario: PMBOK review of local changes
  Given the user has local changes (staged, unstaged, or untracked)
  When /pm-review is invoked
  Then the skill collects the change set and project context, delegates to
       @agent-pm-review:pm-review, and surfaces the structured report

Scenario: Staged-only PM review
  Given the user has staged changes ready to commit
  When /pm-review --staged-only is invoked
  Then only staged changes are gathered and analyzed by the subagent

Scenario: Write report to file
  Given the user wants a persistent record
  When /pm-review --file is invoked
  Then the subagent writes the report to pm-review-YYYY-MM-DD-HHmm.md
       and the skill confirms the path to the user

Scenario: No local changes
  Given the working tree is clean
  When /pm-review is invoked
  Then the user is told "No local changes detected" and the skill exits
       without invoking the subagent

References

• references/pmbok-review-guide.md — Complete PMBOK review framework (read by the subagent, available here for direct human reference).
• references/report-template.md — Exact markdown format for the report.