Content
82%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-engineered procedural skill: fully concrete templates, checklists, and commands with upfront validation, preview, and error-tolerant per-file handling. The body is dense but almost every section carries non-obvious round-trip detail; the residual weaknesses are minor — slight repetition, no post-write verification step, and lookup-grade OKF mapping rules inlined rather than split into a reference file.
Suggestions
Add a post-write verification checkpoint after Step 4 (e.g. re-parse a sample of written pages or re-read frontmatter to confirm round-trip integrity) to close the last workflow-validation gap for this batch-write operation.
Move the OKF reverse-mapping table and derived-key drop rules (Step 4-OKF item 3) into a references/ file (e.g. OKF-MAPPING.md) and keep only the common-case summary inline, shortening the always-loaded body.
Cut the duplicated stub-vs-full and directory-creation statements (intro, line 197, Notes) to a single authoritative location.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with non-obvious protocol detail that earns its tokens (merge-mode matrix, OKF reverse-mapping rules, the folder-note path-resolution rule), and signposting like "Step 2, Step 3 (graph only), and Step 5 are shared; Step 4 forks by source type" keeps it navigable. There is minor over-explanation that could be trimmed — the stub-vs-full distinction is stated three times (intro, Step 4-OKF closing line, Notes) and the directory-creation note is repeated in Notes — so it fits anchor 4 ("Efficient; minor instances of over-explanation that could be trimmed") rather than anchor 5's every-token-earns-its-place, and is well above anchor 3's unnecessary explanations. | 4 / 5 |
Actionability | The guidance is fully concrete and executable for an instruction-only skill: a complete page template with every frontmatter field and conditional blocks, a 7-step merge checklist, exact per-key reverse-mapping rules, source validation criteria, preview formats, a copy-paste bash command (`obsidian-wiki memory sync IMPORT source=... pages_created=<N> ...`), and an exact summary output format. This matches anchor 5 ("Fully executable; copy-paste ready code or commands; specific examples cover the common cases") — anchor 4's "minor gaps" has no referent here. | 5 / 5 |
Workflow Clarity | Steps 1–6 are clearly sequenced (detect/validate → conflict mode → build maps → reconstruct → metadata → summary) with real checkpoints: source validation with "If validation fails, report what's wrong and stop", a preview before writing, per-file error tolerance (unparseable files skipped and counted), mode-based conflict handling, and tracked created/skipped/merged counts. The one gap is that no checkpoint verifies the written pages after the batch write, which is exactly anchor 4's "minor validation gaps"; anchor 5 additionally requires feedback loops for error recovery, which a stop-and-report path only partially provides — so the destructive/batch cap of 3 does not apply, but 5 is not fully earned either. | 4 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, or assets/), and cross-references are clearly signaled and one level deep ("the `Writing Profile Resolution` section in `llm-wiki/SKILL.md`", "Config Resolution Protocol in `llm-wiki/SKILL.md`", "See `.skills/llm-wiki/references/MEMORY.md`"). Sections are well-organized and the shared/forking step structure aids navigation, but the 265-line monolithic body inlines reference-grade material (the ~40-line OKF key-mapping and derived-key drop rules are lookup tables a run only sometimes needs), which keeps it at anchor 4 ("Good structure; most content is appropriately placed; references mostly clear; minor organization gaps") rather than anchor 5's appropriately split content. | 4 / 5 |
Total | 17 / 20 Passed |