Content
81%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 setup procedure with concrete tool calls, copy-ready config blocks, confirmation checkpoints, and an explicit verification phase. Its main weaknesses are mild redundancy around sessionProfile/repository persistence rules and a description/frontmatter trigger gap rather than anything in the body itself.
Suggestions
De-duplicate the sessionProfile/repository persistence guidance, which is stated in the user-level paragraph and again in the following 'Persist sessionProfile explicitly...' paragraph.
Name the exact search tool and arguments for the schema verification step (e.g. bm_search(...)) so it is copy-paste executable like the other steps.
Consider moving the long Interview field-by-field spec into a references file and linking it, keeping the core flow lean in SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is directive and dense with no explanations of concepts Claude already knows, but it has minor redundancy that could be trimmed: the sessionProfile/repository persistence rules are stated twice and the capture-events default is explained in both the Interview and Apply sections. Not a 5 because those repeats cost tokens without adding information. | 4 / 5 |
Actionability | Provides concrete MCP calls (bm_project(command="list"), bm_write with directory and note_type), CLI commands (basic-memory hook status --harness codex --project-dir <repo-root>), and complete copy-adaptable JSON and TOML blocks. Minor gaps remain, e.g. 'Search the primary project for type=schema with page size 10' does not name the exact tool call to use. | 4 / 5 |
Workflow Clarity | Clear sequence (Preconditions, Interview, Apply, Seed Schemas, Verify) with explicit validation steps in Verify and an error-recovery loop ('If any check errors, fix the project ref or hook launcher before finishing'), plus confirmation checkpoints before every write and config edit. Matches the anchor for clear sequence with explicit validation and feedback loops. | 5 / 5 |
Progressive Disclosure | Sections are well organized and the bundle is self-contained (only assets/icon.svg, unreferenced), with external references (<plugin-root>/schemas/) one level deep and clearly signaled. Not a 5 because the long inline interview spec could arguably live in a reference file and the schema filenames (coding-session.md, decision.md, task.md) are named but not linked or listed. | 4 / 5 |
Total | 17 / 20 Passed |