debugging-strategies
Master systematic debugging techniques, profiling tools, and root cause analysis to efficiently track down bugs across any codebase or technology stack. Use when investigating bugs, performance issues, or unexpected behavior.
70
47%
Does it follow best practices?
Impact
84%
0.98xAverage score across 6 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/developer-essentials/skills/debugging-strategies/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 good structural completeness with an explicit 'Use when' clause, but suffers from moderate vagueness in its capability descriptions—listing categories of techniques rather than concrete actions. The trigger terms cover the basics but miss many natural user phrasings, and the 'any codebase or technology stack' framing creates potential overlap with other coding-related skills.
Suggestions
Replace abstract categories with concrete actions, e.g., 'Analyze stack traces, set breakpoints, profile memory and CPU usage, bisect commits to isolate regressions' instead of 'systematic debugging techniques, profiling tools'.
Expand trigger terms in the 'Use when' clause to include common user phrasings like 'error', 'crash', 'not working', 'slow', 'exception', 'stack trace', or 'regression'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (debugging) and some actions ('debugging techniques, profiling tools, root cause analysis'), but these are still fairly abstract categories rather than concrete specific actions like 'set breakpoints, analyze stack traces, inspect memory usage'. | 2 / 3 |
Completeness | Clearly answers both 'what' (systematic debugging techniques, profiling tools, root cause analysis to track down bugs) and 'when' with an explicit 'Use when investigating bugs, performance issues, or unexpected behavior' clause. | 3 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'bugs', 'performance issues', 'unexpected behavior', 'debugging', and 'profiling', but misses common natural variations users might say such as 'error', 'crash', 'stack trace', 'slow', 'broken', 'not working', 'exception'. | 2 / 3 |
Distinctiveness Conflict Risk | While debugging is a recognizable niche, the phrase 'across any codebase or technology stack' is very broad and could overlap with general coding assistance skills, code review skills, or performance optimization skills. The scope is too wide to be clearly distinct. | 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 reads more like a generic debugging textbook chapter than a targeted skill for Claude. It is excessively verbose, explaining well-known concepts (scientific method, rubber duck debugging, 'read error messages') that waste context window tokens. While it contains some useful executable code snippets, the majority of content is abstract checklists and platitudes that don't add actionable value beyond what Claude already knows.
Suggestions
Cut the content by 60-70%: remove generic debugging advice (mindset, common mistakes, best practices, rubber duck debugging) and focus only on concrete tool usage, commands, and code patterns Claude wouldn't already know.
Split language-specific debugging sections (Python, JS/TS, Go) into separate referenced files to improve progressive disclosure and reduce the monolithic structure.
Add explicit validation/verification steps to the debugging workflow, such as 'confirm the fix by running the minimal reproduction case' and 'verify no regression with git stash/apply testing.'
Replace the markdown-in-markdown checklists with concrete, executable examples—e.g., instead of listing 'N+1 queries' as a common culprit, show how to detect one with a specific profiling command and expected output.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose and padded with content Claude already knows. Explains basic concepts like 'the scientific method,' 'rubber duck debugging,' and 'read error messages.' The debugging mindset section, common mistakes, and best practices are all generic advice that wastes tokens. Much of the content is markdown-within-markdown checklists that restate obvious debugging principles. | 1 / 3 |
Actionability | Contains some executable code examples (pdb, Chrome DevTools, git bisect, VS Code launch.json), but much of the skill is abstract checklists and markdown tables rather than concrete, copy-paste-ready guidance. The 'Debugging Patterns by Issue Type' sections are entirely descriptive markdown lists with no executable code. The trace decorator example is useful but the overall ratio of actionable to descriptive content is low. | 2 / 3 |
Workflow Clarity | The four-phase process (Reproduce → Gather Info → Hypothesize → Test) provides a clear sequence, but validation checkpoints are missing. There are no explicit feedback loops for verifying that a fix actually resolves the issue, and the phases themselves are filled with generic checklists rather than concrete verification steps. The git bisect section is the only workflow with clear, sequenced commands. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of text at ~400+ lines with no references to external files and no bundle files to support it. Everything is inlined—language-specific debugging tools, advanced techniques, patterns by issue type—all of which could be split into separate reference files. There's no navigation structure beyond flat headings. | 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 |
|---|---|---|
skill_md_line_count | SKILL.md is long (528 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- wshobson/agents
- Commit
cf6059d
- 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.