performance
Profiling methodology and optimization strategy for performance work. Use when the user asks to "make this faster", "optimize", "profile", "reduce latency", "fix slow", "improve throughput", or when investigating performance regressions.
78
73%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/performance/SKILL.mdQuality
Discovery
89%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 solid skill description with excellent trigger term coverage and a clear 'Use when' clause that makes it easy for Claude to select appropriately. Its main weakness is the somewhat vague 'what' portion—it describes a methodology area rather than listing specific concrete actions the skill enables. Adding explicit capabilities would elevate it further.
Suggestions
Replace 'Profiling methodology and optimization strategy for performance work' with specific concrete actions, e.g., 'Profiles CPU and memory hotspots, identifies bottlenecks, recommends algorithmic and caching optimizations, and benchmarks code changes.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain ('profiling methodology and optimization strategy for performance work') but does not list multiple specific concrete actions. It describes a general area rather than enumerating specific capabilities like 'profile CPU hotspots, analyze memory allocations, benchmark functions, reduce algorithmic complexity.' | 2 / 3 |
Completeness | Clearly answers both 'what' (profiling methodology and optimization strategy for performance work) and 'when' (explicit 'Use when...' clause with multiple trigger scenarios). The when clause is detailed and actionable. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would actually say: 'make this faster', 'optimize', 'profile', 'reduce latency', 'fix slow', 'improve throughput', and 'performance regressions'. These are highly natural phrases covering common variations. | 3 / 3 |
Distinctiveness Conflict Risk | Performance profiling and optimization is a clear niche with distinct trigger terms. The specific phrases like 'profile', 'reduce latency', 'improve throughput' are unlikely to conflict with other skills like general code editing or debugging. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
57%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-structured conceptual guide for performance optimization methodology, with good organization and useful frameworks (trade-off table, bottleneck patterns, output template). Its main weaknesses are the lack of concrete, executable examples — no actual profiling commands or code snippets — and some verbosity in explaining concepts Claude already understands. It reads more like a teaching document than an actionable skill reference.
Suggestions
Add concrete, executable profiling commands for at least 2-3 languages/tools (e.g., `py-spy record -o profile.svg -- python script.py`, `time.perf_counter()` baseline snippet, `console.time()`/`performance.now()` for JS)
Add explicit validation/feedback loop to the Performance Loop: what to do if step 6 shows regression, how to decide whether to keep or revert a change, and a threshold for 'meaningful improvement'
Trim the anti-patterns section and Knuth quote — Claude knows these principles. Replace with a compact 'common mistakes' bullet list and use the saved tokens for executable examples
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is generally well-structured but includes some unnecessary material Claude already knows — the Knuth quote, explanations of common concepts like what caching is, and the trade-off table examples are somewhat verbose. The anti-patterns section restates well-known principles. However, the tables and structured patterns do add value. | 2 / 3 |
Actionability | The skill provides a clear methodology and output format template, but lacks concrete executable code or specific tool commands. There are no actual profiling commands (e.g., `py-spy`, `perf`, `time.perf_counter()` snippets), no benchmark setup examples, and the guidance remains at the conceptual/framework level rather than copy-paste ready. | 2 / 3 |
Workflow Clarity | The Performance Loop provides a clear 6-step sequence, and the profiling strategy has a narrowing approach. However, there are no explicit validation checkpoints or feedback loops — step 6 says 'measure again' but doesn't specify what to do if results regress or are inconclusive. For a methodology involving potentially destructive optimizations, the lack of rollback/verification guidance is a gap. | 2 / 3 |
Progressive Disclosure | The content is well-organized with clear sections progressing from overview to specifics. The 'See Also' section provides one-level-deep references to related skills. Tables break up information effectively. The output format template is appropriately placed at the end. | 3 / 3 |
Total | 9 / 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 | |
- Repository
- tslateman/duet
- Commit
c3b1fc2
- 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.