Conciseness

The Steps, worked example, anti-patterns, and limitations are dense and earn their tokens, but the Overview re-explains blameless postmortem culture — a concept Claude already knows — through four separate SRE quotes ('A postmortem is a written record...', 'Blameless postmortems are a tenet of SRE culture', 'Writing a postmortem is not punishment...', plus 'The blameless framing is load-bearing...'), with the same principle restated again in Step 2 and the Anti-patterns table. Not 4: the redundancy is more than minor trimming — the Overview alone could lose ~15 lines. Not 2: apart from the quote pileup there is no padding or over-explanation.

Actionability

Fully concrete instruction-level guidance: a 9-section skeleton with per-section specs, before/after blameless rewrites ('"Bob deployed without testing" → "The pre-deploy testing didn't cover the null-metadata case"'), four required action-item fields, a storage layout with filename convention, and a completely filled worked example covering the common case. Not 4: the worked example is the copy-paste analog for an instruction-only skill and covers the common path end to end.

Workflow Clarity

Steps 1–5 are clearly sequenced (author → blameless review → action item discipline → approval/closure → storage) with explicit validation checkpoints (reviewer sign-off, action items in tracker, 'closed only when all action items ship') and feedback loops (Step 2's reject-and-rewrite patterns for blame language). Not 4: checkpoints are explicit rather than 'most present' — no significant validation gaps.

Progressive Disclosure

Structure is appropriate: SKILL.md is the overview with a compact inline section list, while the full per-incident template lives in the clearly signaled, one-level-deep references/post-mortem-document-template.md (verified to exist and match what the body claims). Anti-patterns, worked example, and limitations — which belong inline — are inline. Not 4: navigation is clean with no buried references and nothing that should be external is inlined.

Specificity

The description lists multiple specific concrete actions with comprehensive coverage: 'captures the timeline (chronological event sequence with sources), root cause analysis (what + why, not who), impact (users / revenue / SLO debt), action items (with owners + due dates + measurable success criteria), and "what went well"' — each deliverable is named with its defining detail. Not 4: coverage is comprehensive with no material gaps; it exceeds the 'minor gaps' of the 4 anchor.

Completeness

Explicitly answers both questions: what — 'produces a blameless post-mortem from an incident' with the five content areas enumerated; when — 'Use after every user-visible incident, not just severe ones', a concrete trigger clause. Not 4: the 'when' is explicit and specific rather than 'could be more explicit'.

Trigger Term Quality

Good natural keyword coverage — 'post-mortem', 'blameless', 'incident', 'root cause', 'impact', 'action items', 'user-visible incident' — terms a user would naturally say. Not 5: misses common synonyms like the one-word 'postmortem', 'retrospective'/'retro', and 'incident review'/'SEV'. Not 3: coverage goes well beyond 'some relevant keywords', including the canonical phrasings.

Distinctiveness Conflict Risk

Clear niche — blameless incident post-mortem authoring — with distinct triggers ('after every user-visible incident'). Minimal conflict risk; only slight adjacency to generic retrospective/write-up skills. Not 4: the trigger set and deliverables are specific enough that mis-triggering is unlikely.