aaronguo/muse
Conversational writing topic discovery. Combines personal context (journals, notes, past writing), optional Signal DB intelligence, and web research to surface timely, authentic topic ideas for any writing format.
90
90%
Does it follow best practices?
Impact
Pending
Average score across 4 eval scenarios
Risky
Do not use without reviewing
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 strong skill description that excels across all dimensions. It clearly articulates specific capabilities (reading personal content, accessing trends, web search), provides excellent trigger term coverage in multiple languages, and includes an explicit 'Use when' clause. The only minor issue is the use of second person 'your' which slightly deviates from the preferred third person voice, but this doesn't significantly impact functionality.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'Reads your journals, notes, and past work', 'taps into an optional Signal DB for real-time trends', 'searches the web', 'surfaces topic ideas'. These are specific, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both what (reads journals/notes, accesses Signal DB, searches web, surfaces topic ideas) AND when (explicit 'Use when' clause with multiple trigger phrases). Fully complete. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including 'find a topic', 'what should I write about', 'inspire me', 'topic ideas', 'help me pick a topic', plus Chinese variants '选题' and '写什么好'. These are phrases users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Clear niche focused on writing topic ideation with distinct triggers. The combination of personal content analysis, trend database, and web search for topic generation is specific enough to avoid conflicts with general writing or research 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 skill with excellent actionability and workflow clarity. The multi-step process is clearly sequenced with explicit decision points, fallbacks, and validation criteria. The main weaknesses are moderate verbosity (philosophical framing and repeated principles) and a somewhat monolithic structure that could benefit from extracting detailed sub-workflows to reference files.
Suggestions
Remove or condense the introductory paragraph and 'Muse's Principles' section - these concepts are already embedded in the workflow steps
Consider extracting Step 6 (Go Deeper) sub-steps to a separate reference file like 'references/topic-refinement.md' to reduce main file length
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is well-written but includes some unnecessary prose and philosophical framing ('Your muse knows what you've been thinking about...', 'The output is not a plan. It's a direction.'). The principles section at the end repeats concepts already clear from the workflow. Could be tightened by ~20%. | 2 / 3 |
Actionability | Provides fully executable curl commands for Signal DB queries, concrete bash commands for file checks, specific query patterns for web searches, and detailed output formats. The guidance is copy-paste ready with clear variable substitution patterns. | 3 / 3 |
Workflow Clarity | Seven clearly sequenced steps with explicit decision points (EXTEND.md found/not found, Signal DB enabled/disabled, content paths empty). Includes validation checkpoints ('When to move on' criteria in Step 6), fallback behaviors, and clear hand-off conditions. The workflow handles edge cases explicitly. | 3 / 3 |
Progressive Disclosure | References external files appropriately (EXTEND.md, first-time-setup.md) but the main skill file is quite long (~200 lines). Step 6's sub-steps could potentially be extracted to a separate reference. The structure is good but the content density in a single file is high. | 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