Content
88%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 diagnostic procedure: exact commands and tool calls, a fixed output template, and thorough fallback handling for every failure mode. The only minor improvements are trimming a couple of verbose spots and deciding whether the config-key/default enumeration belongs inline or in a small reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and assumes Claude's competence — no background concept explanations, just procedure — but a few spots could be tightened: step 2's config-precedence prose ('do not trim it, expand ~, or treat an empty value as unset') runs long, and the instruction to pass the project's external ID as `project` is stated twice. This is 'efficient; minor instances that could be trimmed' rather than the every-token-earns-its-place top anchor. | 4 / 5 |
Actionability | Fully executable throughout: exact launcher commands with fallbacks (`basic-memory --version`, then `bm --version`, then `uvx --prerelease=allow basic-memory --version`), a precise hook command (`basic-memory hook status --harness claude --project-dir <project-root>`), concrete MCP calls with argument syntax (`bm_search` with `filters={"type": "session"}`, `page_size` 5), and a copy-paste-ready output template. As an instruction-only skill, its guidance is concrete and complete with no gaps. | 5 / 5 |
Workflow Clarity | A clear numbered gather sequence (CLI reachability → configuration → hook health → checkpoints → tasks) followed by a fixed presentation template, with explicit error-recovery branches: unavailable launchers degrade gracefully ('mark these fields unavailable and continue'), config mismatches are surfaced, undetermined values are written as '—' rather than failing the report, and empty-state/warning rules are spelled out. The read-only diagnostic involves no destructive or batch operations, so no validation cap applies. | 5 / 5 |
Progressive Disclosure | The skill is self-contained with no bundle files and clean two-section organization, and everything inlined is needed at invocation time. It sits just above the rubric's under-50-line simple-skill threshold, and the inline enumeration of `basicMemory` config keys with defaults is borderline reference material, so this fits 'good structure; most content appropriately placed; minor organization gaps' rather than a flawless 5. | 4 / 5 |
Total | 18 / 20 Passed |