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 well-structured, highly actionable skill body with executable commands, an explicit pre-scaffold validation checklist, and clear external references. The main room for improvement is splitting detailed reference material into bundle files and trimming some explanatory prose.
Suggestions
Move the YAML key-fields reference and the client character-limit table into a references/ bundle file, keeping SKILL.md as a concise overview with a clearly signaled link.
Tighten the 'Enabling or renaming a tool' prose into a short checklist of deploy-sequencing rules to reduce token overhead.
Add a couple of trigger-term synonyms to the description (e.g. 'MCP server', 'agent tools') to broaden natural-phrase coverage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense and technical with no basic-concept padding; every section earns its place, though some explanatory prose (e.g. the enabling/renaming deployment narrative) could be trimmed slightly, placing it above the midpoint but not fully lean. | 4 / 5 |
Actionability | Provides copy-paste executable commands ('pnpm --filter=@posthog/mcp run scaffold-yaml …', 'hogli build:openapi', 'lint-tool-names', '--sync-all') and a concrete YAML example with real fields and file paths, covering the common cases. | 5 / 5 |
Workflow Clarity | Sequenced quick-workflow (steps 1–4) is paired with an explicit pre-scaffold verification checklist (serializer types, @extend_schema, @validated_request) and error-recovery guidance ('empty/wrong schema → fix on the Django side'), with sync-all noted as non-destructive so no cap applies. | 5 / 5 |
Progressive Disclosure | Clear section headers and well-signaled one-level-deep external references (full guide, type-system guide, improving-drf-endpoints skill), but no bundle files exist so detailed reference material (the YAML field reference and client-limit table) is inlined rather than split out, leaving minor organization gaps. | 4 / 5 |
Total | 18 / 20 Passed |