thinking-deeply
Engages structured analysis to explore multiple perspectives and context dependencies before responding. Use when users ask confirmation-seeking questions, make leading statements, request binary choices, or when feeling inclined to quickly agree or disagree without thorough consideration.
49
53%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./think_deeply/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 has a well-structured format with an explicit 'Use when' clause that clearly delineates trigger conditions, which is its strongest aspect. However, the capability description is somewhat abstract and process-oriented rather than listing concrete actions, and the trigger terms lean toward meta-cognitive language rather than natural user phrasings. The skill's niche is moderately distinct but could benefit from sharper differentiation.
Suggestions
Replace abstract process language with concrete actions, e.g., 'Identifies assumptions, generates counterarguments, weighs trade-offs, and surfaces hidden context dependencies' instead of 'engages structured analysis to explore multiple perspectives.'
Add more natural user-facing trigger terms to the 'Use when' clause, such as 'is this right?', 'should I do X?', 'yes or no', 'do you agree', 'what's the best option' — phrases users would actually type.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a domain ('structured analysis', 'multiple perspectives', 'context dependencies') and describes a general approach, but the actions are somewhat abstract—'engages structured analysis' and 'explore multiple perspectives' are process-oriented rather than listing concrete, specific actions like 'generates pro/con lists' or 'identifies assumptions.' | 2 / 3 |
Completeness | Clearly answers both 'what' (engages structured analysis to explore multiple perspectives and context dependencies) and 'when' (explicit 'Use when' clause covering confirmation-seeking questions, leading statements, binary choices, and inclination to quickly agree/disagree). The 'Use when' clause is explicit and detailed. | 3 / 3 |
Trigger Term Quality | Includes some useful trigger terms like 'confirmation-seeking questions', 'leading statements', 'binary choices', which are somewhat natural. However, these are more analytical/meta-cognitive terms than what users would naturally say—users are unlikely to say 'I have a confirmation-seeking question.' Missing more natural phrasings like 'yes or no question', 'should I do X', 'is this right', 'what do you think.' | 2 / 3 |
Distinctiveness Conflict Risk | The skill occupies a somewhat unique niche around critical thinking and avoiding hasty agreement, but 'structured analysis' and 'explore multiple perspectives' could overlap with general reasoning, debate, or analysis skills. The trigger conditions help differentiate it, but the core capability description is broad enough to potentially conflict with other analytical skills. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
39%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is well-structured in terms of workflow but severely undermined by verbosity. The three examples are repetitive and demonstrate the same pattern, consuming significant token budget. The content over-explains a reasoning approach that Claude is already capable of, and would benefit greatly from condensing to ~50 lines with examples moved to a separate file.
Suggestions
Reduce to one example (or move examples to a separate EXAMPLES.md) since all three demonstrate the identical pattern of reframe → perspectives → context → recommendation.
Remove the 'Purpose', 'Remember', and 'Success Criteria' sections which restate the same idea three times—consolidate into a single brief sentence at the top.
Trim the 'When This Skill Activates' list to 2-3 items since items like 'confirmation-seeking questions' and 'quick validation requests' overlap significantly.
Consider linking to a separate file for the detailed structured response format template (step 4) to keep the main skill lean.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~200+ lines. The three full examples are largely repetitive, each demonstrating the same pattern (reframe → perspectives → context dependencies → recommendation). One example would suffice. The anti-patterns section, success criteria, and 'Remember' section all restate the same core idea. Much of this explains reasoning patterns Claude already knows well. | 1 / 3 |
Actionability | The skill provides a clear structured protocol (pause, reframe, map, respond) with concrete response templates and before/after examples. However, this is an instruction-only skill with no code or commands, and the guidance is somewhat generic—the structured response format reads more like a rigid template than flexible, actionable guidance for diverse situations. | 2 / 3 |
Workflow Clarity | The 5-step protocol is clearly sequenced (Pause → Reframe → Map → Respond → Anti-patterns). Each step has explicit sub-steps. The structured response format (a through d) provides a clear template. For a non-destructive reasoning skill, this level of workflow clarity is appropriate and well-organized. | 3 / 3 |
Progressive Disclosure | The content is a monolithic wall of text with no references to external files. The three lengthy examples could easily be in a separate EXAMPLES.md. The trigger conditions, anti-patterns, and success criteria sections add bulk that could be condensed or split out. Everything is inline with no navigation structure. | 1 / 3 |
Total | 7 / 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.
- Repository
- brunoasm/my_claude_skills
- Commit
2a19b68
- Reviewed
Table of Contents
Is this your skill?
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.