exa-search
AI-powered web search via Exa with content extraction. Use when user says "exa search", "web search with content", "find similar pages", or needs broad web results beyond academic databases (arXiv, Semantic Scholar).
80
77%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/exa-search/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 description with excellent trigger terms and completeness, including an explicit 'Use when' clause and clear differentiation from related academic search skills. The main weakness is that the 'what' portion could be more specific about the concrete actions performed (e.g., summarizing results, extracting full page content, finding semantically similar URLs).
Suggestions
Expand the capability list with more concrete actions, e.g., 'Searches the web via Exa API, extracts full page content, finds semantically similar pages, and filters by domain or date.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (web search via Exa) and mentions 'content extraction' and 'find similar pages', but doesn't list multiple concrete actions comprehensively—e.g., it doesn't specify what kinds of content extraction or what outputs are produced. | 2 / 3 |
Completeness | Clearly answers both 'what' (AI-powered web search via Exa with content extraction) and 'when' (explicit 'Use when' clause with specific trigger phrases and a scope boundary distinguishing it from academic database skills). | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms: 'exa search', 'web search with content', 'find similar pages', and differentiates from academic databases (arXiv, Semantic Scholar). These are terms users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Explicitly distinguishes itself from academic database tools (arXiv, Semantic Scholar) and uses the unique trigger 'exa search'. The scope boundary ('broad web results beyond academic databases') makes it clearly distinguishable from related search skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a functional skill with strong actionability — concrete commands, clear argument specifications, and multiple usage examples. Its main weaknesses are moderate verbosity (the comparison table, setup basics, and extensive argument list could be trimmed) and missing validation/error-handling checkpoints in the workflow, particularly around API failures and empty result sets. The wiki integration in Step 6 is a nice touch but adds complexity without corresponding error recovery.
Suggestions
Add validation checkpoints after Step 3 (check for API errors, empty results, missing API key at runtime) to improve workflow robustness.
Remove or significantly trim the comparison table and setup section — Claude knows how to pip install and export env vars; the description field already handles routing.
Consider moving the full argument specification list to a separate reference file and keeping only the most common options inline to improve conciseness and progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary content like the comparison table with other skills (Claude can be told when to use it via the description), and the setup section explains basic pip install and env var export which Claude already knows. However, the core workflow and argument parsing sections are reasonably efficient. | 2 / 3 |
Actionability | The skill provides fully executable bash commands for every operation mode (search, find-similar, get-contents), concrete argument parsing specifications, and specific examples with real flags. The Step 6 wiki integration includes conditional logic with executable commands. | 3 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced and includes a script-not-found check, but it lacks validation checkpoints after search execution (e.g., checking for empty results, API errors, or malformed responses). The wiki integration step has conditional logic but no error recovery if ingest_paper fails. | 2 / 3 |
Progressive Disclosure | The skill references `shared-references/integration-contract.md` for wiki details which is good progressive disclosure, but the main body is quite long (~130 lines) with detailed argument parsing and multiple workflow steps that could benefit from being split. No bundle files are provided to verify referenced paths exist. | 2 / 3 |
Total | 9 / 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
2028ac4
- 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.