exa-advanced-troubleshooting
Apply advanced debugging techniques for hard-to-diagnose Exa issues. Use when standard troubleshooting fails, investigating latency spikes, or preparing evidence bundles for Exa support escalation. Trigger with phrases like "exa hard bug", "exa mystery error", "exa deep debug", "difficult exa issue", "exa latency spike".
85
83%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
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 well-structured description with strong completeness and distinctiveness, clearly positioning itself as an advanced/escalation debugging skill for Exa. Its main weakness is that the core capabilities are somewhat vague—'advanced debugging techniques' doesn't enumerate specific concrete actions beyond latency spike investigation and evidence bundle preparation. The trigger terms are well-chosen and natural.
Suggestions
Replace 'apply advanced debugging techniques' with specific concrete actions, e.g., 'Analyze Exa request traces, profile query performance, inspect API response anomalies, and correlate error patterns across logs.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | It names the domain (Exa debugging) and mentions some actions like 'investigating latency spikes' and 'preparing evidence bundles for Exa support escalation,' but the core action 'apply advanced debugging techniques' is vague and doesn't list concrete specific actions (e.g., analyzing logs, tracing requests, profiling queries). | 2 / 3 |
Completeness | Clearly answers both 'what' (advanced debugging for hard-to-diagnose Exa issues, investigating latency spikes, preparing evidence bundles) and 'when' (when standard troubleshooting fails, with explicit trigger phrases listed). Has an explicit 'Use when' clause and a 'Trigger with' clause. | 3 / 3 |
Trigger Term Quality | Includes a good set of natural trigger phrases like 'exa hard bug', 'exa mystery error', 'exa deep debug', 'difficult exa issue', 'exa latency spike' that users would plausibly say. Also includes contextual triggers like 'standard troubleshooting fails' and 'support escalation.' | 3 / 3 |
Distinctiveness Conflict Risk | Clearly positioned as an escalation-level debugging skill specifically for Exa, distinct from standard Exa troubleshooting or general debugging skills. The 'when standard troubleshooting fails' qualifier and Exa-specific trigger terms create a clear niche. | 3 / 3 |
Total | 11 / 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 strong, highly actionable skill with executable diagnostic code and a clear multi-step debugging workflow. Its main weakness is verbosity — the extensive inline code blocks make the skill quite long when some of the diagnostic functions could be referenced from separate files. The error handling table and escalation template add significant practical value.
Suggestions
Consider extracting the longer diagnostic functions (diagnoseExa, profileLatency, debugContentRetrieval) into a separate referenced script file, keeping only usage examples inline in SKILL.md.
Tighten the diagnostic code by removing some of the repetitive try/catch/push patterns — Claude can generalize from one or two examples rather than five nearly identical blocks.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The code examples are thorough but quite lengthy. The full diagnostic function (~80 lines) could be more concise since Claude can infer boilerplate patterns. However, each section does serve a distinct purpose and there's minimal prose padding. | 2 / 3 |
Actionability | All code is fully executable TypeScript with proper imports, typed interfaces, and concrete API calls. The diagnostic function, latency profiler, content retrieval debugger, and escalation template are all copy-paste ready with specific Exa SDK methods. | 3 / 3 |
Workflow Clarity | The four steps form a clear diagnostic sequence: layer-by-layer diagnosis → latency profiling → content retrieval debugging → support escalation. Each step builds on the previous, with early-exit validation (e.g., stopping if network fails, stopping if auth fails) and the error handling table provides a feedback loop for common issues. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear sections, but the inline code is quite long (~150+ lines of TypeScript) which could benefit from being split into separate reference files. The single reference to 'exa-load-scale' and external docs is appropriate, but the main file is heavy for an overview. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
3a2d27d
- 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.