Content
93%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A tight, executable diagnostic workflow that respects the context budget and pushes detail to one-level-deep references. The only minor gap is the absence of an explicit validate/retry loop for the "Unknown" verdict path, which keeps workflow clarity just below the top anchor.
Suggestions
Add a brief feedback step for the "Unknown" verdict (e.g., query the session's full event timeline and re-evaluate against the signal list) to close the workflow's only validation gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence — it never explains what PostHog, an SDK, or an ad blocker is; every table, SQL query, and verdict entry earns its place. | 5 / 5 |
Actionability | Fully executable guidance: copy-paste `posthog:session-recording-get` call, exact-property SQL queries, and a verdict→recommendation table covering the common diagnostic cases. | 5 / 5 |
Workflow Clarity | A clear 5-step sequenced workflow with branching at Step 1 (exists vs 404) and prioritized verdicts at Step 3; it lacks an explicit re-query/retry feedback loop for the "Unknown" verdict, which is a minor validation gap. | 4 / 5 |
Progressive Disclosure | SKILL.md is a concise overview with clearly signaled, one-level-deep markdown links to three real reference files (diagnostic-signals.md, diagnosis-logic.md, examples.md); detail is appropriately split and easy to navigate. | 5 / 5 |
Total | 19 / 20 Passed |