crocoder-dev/opencode-sce-context-sync
Sync Shared Context Engineering context files with implemented code changes.
93
93%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-crafted skill description that excels across all dimensions. It opens with an explicit 'Use when' clause containing multiple natural trigger phrases, then clearly describes the concrete actions performed and the specific artifacts affected. The focus on AI memory/context files under a specific directory gives it a distinct identity that minimizes conflict risk.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: scans modified code, classifies change significance, updates/verifies Markdown context files. Also enumerates specific file types: overview, architecture, glossary, patterns, context-map, and domain files under `context/`. | 3 / 3 |
Completeness | Clearly answers both 'what' (scans modified code, classifies change significance, updates Markdown context files) and 'when' (explicit 'Use when' clause covering updating docs after code changes, syncing docs, refreshing context, keeping memory files accurate). | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would say: 'update project documentation', 'sync docs with code', 'refresh project context', 'keep AI memory files accurate', 'implementation task'. These cover multiple natural phrasings a user might use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: updating AI memory/context Markdown files under `context/` to reflect code changes. The specific mention of durable AI memory, context files, and the `context/` directory makes it unlikely to conflict with general documentation or code skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted process skill with strong actionability and workflow clarity. The step-by-step sync pass, classification table with concrete examples, and before/after illustration make it easy for Claude to follow. Minor weaknesses include some redundancy between sections and the fact that all content lives in a single file rather than leveraging progressive disclosure through supporting bundle files.
Suggestions
Consider extracting the Classification Reference table and Domain File Creation Policy into separate bundle files (e.g., classification-reference.md, domain-file-policy.md) to improve progressive disclosure and reduce the main file's length.
Remove the slight redundancy between the 'Root context significance gating' section and the 'Classification Reference' table—one could reference the other rather than restating similar criteria.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some redundancy—the classification reference table partially restates the gating section, and some points (like 'ensure feature existence') are repeated across the workflow and final-task requirement. Overall it respects Claude's intelligence but could be tightened. | 2 / 3 |
Actionability | The skill provides a concrete step-by-step workflow, a clear classification table with real examples, a before/after example showing exactly what files to create/update and their content, and specific quality constraints (250-line limit, relative paths, Mermaid diagrams). This is highly actionable guidance. | 3 / 3 |
Workflow Clarity | The 9-step workflow is clearly sequenced with explicit decision points (classify → verify → edit or skip → domain files → context-map → glossary → final check). The final check step serves as a validation checkpoint, and the verify-only vs. root-edit gating provides a clear decision framework. The final-task requirement adds an additional verification loop. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and internal anchor links (e.g., [Classification Reference], [Domain File Policy]), but everything is in a single file. Given the length (~100 lines of substantive content) and the classification table, domain file policy, and quality constraints, some of this could be split into referenced files. However, no bundle files exist to support such splitting, and the internal anchoring partially compensates. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
Reviewed
Table of Contents