Content
90%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, highly actionable prod-error hunting workflow with executable commands and a clear sequence. The main gap is the absence of explicit error-recovery feedback loops in the early investigative steps.
Suggestions
Add a short validate→retry loop for the Sentry/log correlation (e.g. 'if firstSeen yields no matching commit, widen --since to 5d and re-check').
Spell out a concrete fallback when the asar-extracted server.js does not match the stack's source lines (e.g. check release tag vs. local HEAD).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and dense — assumes Claude knows OTLP, Sentry, asar, and undici without explaining them, and every line (log paths, the tally one-liner, the blind-spot caveat) earns its place. | 5 / 5 |
Actionability | Provides copy-paste-ready commands — the rg|python3 5xx tally, the asar extract-file command, exact Sentry tags and org/project/region — covering the common cases concretely. | 5 / 5 |
Workflow Clarity | A clear five-step sequence with checkpoints ('Confirm the shipped code matches source before reasoning about it', the Bun/undici repro gate), but lacks explicit validate→fix→retry recovery loops for the log and Sentry steps. | 4 / 5 |
Progressive Disclosure | Well organized into five numbered sections with nothing bloating the file that belongs elsewhere; no bundle files exist, so all guidance is appropriately inline with clear navigation. | 4 / 5 |
Total | 18 / 20 Passed |