perf-review

name	perf-review
description	Gather evidence and draft a review narrative — a self-review / brag-doc for yourself, or a review of someone else — from 1-on-1s, weekly mentions, projects, and incidents, without inventing ratings. Use when the user wants to prepare or update a performance review or self-review. Pass the person's name as args, or none for a self-review.

Prepare a performance review. The subject is the person named in args; with no args, the subject is the vault owner (a self-review / brag-doc).

Resolve role first: read owner.role from .agents/config.yaml (default em) and open the role pack at .agents/roles/{role}.md. Use its {{review_dimensions}} and perf-review skill note. If no subject was given and the role's note defaults to self-review (e.g. ic, pm), proceed as a self-review; otherwise resolve the subject name to a slug via people:.

This is an evidence-gathering and narrative-drafting workflow. Do NOT make up ratings — surface evidence and let the reader draw conclusions.

Evidence sources depend on the subject:

• Self-review (subject = owner): the owner has no profile/1-on-1s. Start with the brag log ({paths.reports}/brag-log.md) if it exists — it's the richest, pre-curated source. Then supplement from the owner's own weekly files (wins, deliverables, #win/#risk), projects where owner == owner.slug, decisions authored, and incidents owned. Skip the @mention and profile steps below.
• Someone else (subject ≠ owner): use all steps below.

1. Read 02-people/profiles/{slug}.md — note role, level, growth_areas, strengths, current cycle expectations. (others only)
2. Determine the current review cycle from context (check 02-people/reviews/ for existing cycle directories). Use the most recent incomplete cycle.
3. Read all 1-on-1 files for this person in 02-people/one-on-ones/{slug}/ from the review period. Extract feedback given (#feedback), wins (#win), concerns/risks (#risk, #blocked), and growth progress. (others only)
4. Scan weekly files in 01-weekly/ — for others, all mentions of @{FirstName} (note patterns: praise, concern, delegation, escalation); for a self-review, your own entries (wins, deliverables, recurring themes).
5. Check 03-projects/ for projects the subject owned or contributed to: match the slug in each hub's owner/team_members (for a self-review, owner == owner.slug), plus any project whose linked items name the subject. Note outcomes.
6. Check 05-incidents/ for incidents the subject was involved in (owner or contributor) — note the response.
7. Determine the output path and create from _templates/performance-review.md if absent:
   • Others: 02-people/reviews/{cycle}/{slug}.md with person: {slug}.
   • Self-review: 02-people/reviews/{cycle}/{owner.slug}.md with person: {owner.slug}.
   • Both set cycle: and status: draft.
8. Populate or append to the review file:
   • ## Evidence Log — bulleted list of raw evidence per source (the brag log for a self-review, weeklies, projects, incidents, and 1-on-1s for others), each with date and source file reference
   • ## Competency Notes — brief notes per the role pack's {{review_dimensions}}
   • ## Draft Narrative — 3–5 sentences summarizing the period, highlighting standout moments

Output a summary of evidence found and confirm the file path.