Comprehensive documentation synchronization - scan local git changes and propagate updates to ALL design docs, task lists, specs, diagrams, and planning artifacts. Use when finishing a feature, after merging, or when design docs are out of date.
64
47%
Does it follow best practices?
Impact
93%
1.03xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.github/skills/doc-sync-all/SKILL.mdQuality
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description does well at answering both 'what' and 'when' with explicit trigger scenarios, which is its strongest aspect. However, the actions described are somewhat high-level ('scan' and 'propagate updates') rather than listing specific concrete operations, and the trigger terms could better cover natural user language variations. The broad scope of artifact types mentioned creates some overlap risk with more specialized documentation or planning skills.
Suggestions
Add more natural trigger terms users might say, such as 'sync docs', 'update documentation', 'docs are stale', 'docs out of sync with code', or 'refresh specs'.
Make capabilities more specific by listing concrete actions like 'update API specs from code changes', 'mark completed tasks in task lists', 'update architecture diagrams to reflect new components'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (documentation synchronization) and some actions ('scan local git changes', 'propagate updates'), but the list of targets ('design docs, task lists, specs, diagrams, and planning artifacts') is more of a category enumeration than specific concrete actions like 'update sequence diagrams' or 'reconcile task completion status'. | 2 / 3 |
Completeness | Clearly answers both 'what' (scan git changes and propagate updates to design docs, task lists, specs, diagrams, planning artifacts) and 'when' (when finishing a feature, after merging, or when design docs are out of date) with explicit trigger scenarios. | 3 / 3 |
Trigger Term Quality | Includes some natural terms like 'design docs', 'task lists', 'specs', 'out of date', 'merging', and 'finishing a feature', but misses common variations users might say such as 'sync docs', 'update documentation', 'docs are stale', 'README', or 'changelog'. The phrase 'documentation synchronization' is somewhat formal. | 2 / 3 |
Distinctiveness Conflict Risk | The scope is fairly distinct (syncing documentation with code changes), but the broad mention of 'ALL design docs, task lists, specs, diagrams, and planning artifacts' could overlap with skills focused on individual documentation tasks, project management, or diagram generation. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is overly verbose and project-specific, reading more like internal documentation for a particular repository than a generalizable skill. While it demonstrates thorough thinking about documentation synchronization workflows, it buries actionable guidance under excessive structural scaffolding, project-specific examples (specific task IDs, FR numbers, file paths), and content Claude could infer. The monolithic format with no progressive disclosure makes it expensive in token budget for what it delivers.
Suggestions
Cut content by at least 50%: remove project-specific examples (T094, FR-021, specific file trees), 'When to Use' and 'Sample Invocation Phrases' sections, and glob patterns that Claude can infer. Focus on the synchronization algorithm, not the inventory.
Split into SKILL.md (overview + core workflow) and referenced sub-files like SYNC-RULES.md (the per-document-type sync rules) and TEMPLATES.md (the manifest and report templates).
Replace the pseudocode validation pattern with a concrete, executable script or at minimum a precise algorithm Claude can follow mechanically, and add explicit error recovery steps (e.g., 'if cross-reference is broken, do X').
Generalize the content: instead of hardcoding specific document names and structures, describe the pattern-matching approach for discovering which docs need updates from which code changes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~300+ lines. Much of the content is project-specific scaffolding (exact file trees, specific task IDs like T094, specific FR numbers) that won't generalize. Explains obvious concepts like what git status does, includes lengthy tables mapping change types to docs that Claude could infer, and repeats patterns across phases. The 'Sample Invocation Phrases' and 'When to Use This Skill' sections explain things Claude already knows. | 1 / 3 |
Actionability | Provides concrete bash commands for git analysis and specific tool names (replace_string_in_file), but the core synchronization logic is procedural description rather than executable code. The 'validation pattern' is pseudocode, not a real script. Most guidance is 'check X, then update Y' without concrete implementation of how to actually perform the matching/updating. | 2 / 3 |
Workflow Clarity | Has a clear 5-phase structure with numbered steps, and Phase 4 Step 5 includes verification. However, the verification step is generic ('re-read modified files to confirm changes') without concrete validation commands or explicit error recovery loops. For a skill involving batch document modifications, the feedback loop is too vague to cap above 2. | 2 / 3 |
Progressive Disclosure | Monolithic wall of text with no references to external files. All content is inline in a single massive document — the file tree examples, sync rules for every document type, glob patterns, sample phrases, and checklists could all be split into referenced sub-documents. The document tries to be both an overview and a complete reference simultaneously. | 1 / 3 |
Total | 6 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
1e34e1c
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.