analyze-crash
Stack Trace Crash Analysis for dd-trace-dotnet
40
38%
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 ./.claude/skills/analyze-crash/SKILL.mdQuality
Discovery
22%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 description reads more like a title than a functional skill description. It identifies the domain (crash analysis for dd-trace-dotnet) but fails to enumerate specific capabilities or provide explicit trigger guidance. Without concrete actions and a 'Use when...' clause, Claude would struggle to reliably select this skill from a larger pool.
Suggestions
Add specific concrete actions such as 'Analyzes native and managed stack traces, identifies crash root causes, maps symbols to source code, and diagnoses common failure patterns in dd-trace-dotnet.'
Add an explicit 'Use when...' clause, e.g., 'Use when the user shares a crash dump, stack trace, exception, or segfault related to dd-trace-dotnet or the Datadog .NET tracer.'
Include natural trigger term variations like 'exception', 'error log', 'crash dump', 'Datadog .NET tracer', 'segfault', 'access violation', and '.NET profiler' to improve matching.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a domain ('Stack Trace Crash Analysis') and a specific project ('dd-trace-dotnet'), but does not list any concrete actions like 'parses stack traces', 'identifies root causes', or 'suggests fixes'. It's essentially a title, not a description of capabilities. | 1 / 3 |
Completeness | The description only vaguely addresses 'what' (crash analysis) and completely lacks a 'when' clause. There is no explicit trigger guidance such as 'Use when...' which per the rubric should cap completeness at 2 at best, but the 'what' is also too weak to merit a 2. | 1 / 3 |
Trigger Term Quality | Contains some useful trigger terms like 'stack trace', 'crash', and 'dd-trace-dotnet' that users might naturally mention. However, it misses common variations like 'exception', 'error', 'Datadog', '.NET tracer', 'crash dump', 'segfault', or 'access violation'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'dd-trace-dotnet' provides some specificity to a particular project, which helps distinguish it. However, 'Stack Trace Crash Analysis' is broad enough that it could overlap with general debugging or error analysis skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides excellent actionable guidance with a well-structured multi-phase workflow for crash analysis, but it is severely over-long and monolithic. The content would benefit greatly from splitting reference material (output template, classification categories, GitHub link rules) into separate files and trimming explanations that Claude can infer from context. The verbosity undermines what is otherwise a very thorough and well-sequenced analysis procedure.
Suggestions
Extract the full output markdown template into a separate TEMPLATE.md file and reference it from the main skill, reducing the main file by ~40 lines.
Remove or drastically shorten the GitHub Link Generation section — Claude can construct URLs from a single example pattern without 15+ lines of explanation about path stripping.
Move the classification categories (CLR Runtime, dd-trace-dotnet Native, etc.) into a separate CATEGORIES.md reference file to keep the main workflow lean.
Trim redundant guidelines — e.g., 'Be concise but thorough' and 'Focus on triage and understanding' repeat what's already conveyed by the workflow structure itself.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~200+ lines. It over-explains concepts Claude already knows (how to parse stack traces, how to read code, how to classify frames). The GitHub link generation section alone is excessively detailed with multiple examples of path stripping that Claude can infer. The output template is a massive markdown block that could be much more compact. | 1 / 3 |
Actionability | The skill provides highly concrete, executable guidance: specific git commands, exact glob patterns for file lookup, precise classification categories with namespace patterns, exact GitHub URL construction rules, and specific shell commands for directory creation on different platforms. The workflow steps are specific and copy-paste ready. | 3 / 3 |
Workflow Clarity | The 5-phase workflow is clearly sequenced (Parse → Locate → Extract → Reconstruct → Related Code) with explicit steps within each phase. It includes graceful degradation ('If not found, use Glob tool'), clear ordering constraints ('Focus on the crashed thread'), and handles edge cases (missing files, path variations, multiple threads). | 3 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no bundle files or external references. The output template, GitHub link generation rules, classification categories, and platform-specific commands could all be split into separate reference files. Everything is inlined into a single massive document. | 1 / 3 |
Total | 8 / 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
- DataDog/dd-trace-dotnet
- Commit
6249bb4
- 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.