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).

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, mostly executable skill body with clear workflow sequencing and validation checkpoints for the fragile script-resolution and batch wiki-ingest steps. Weaknesses are minor: duplicated resolution logic and a pseudocode wiki loop that breaks the otherwise copy-paste-ready pattern.

Suggestions

Factor the shared script-resolution cascade (Steps 2 and 6) into the referenced integration-contract.md so the body only shows the call, removing the duplicated bash blocks.

Rewrite Step 6's wiki ingest as real executable bash (a loop over results) instead of mixing English control flow ('for each research-paper hit in results:') with shell syntax.

Tighten the Role & Positioning table and the Key Rules rationale ('good balance of speed and context') to reduce tokens that assume Claude needs the motivation explained.

Dimension	Reasoning	Score
Conciseness	Largely lean and specific to the Exa SDK / ARIS integration with executable commands, but the Role & Positioning comparison table and the duplicated script-resolution cascade in Steps 2 and 6 could be trimmed.	4 / 5
Actionability	Search, find-similar, and get-contents examples are copy-paste executable bash, but Step 6's wiki loop mixes English pseudocode ('for each research-paper hit in results:') with bash, leaving a minor gap.	4 / 5
Workflow Clarity	Steps 1-6 are clearly sequenced with explicit validation (Step 2 fail-with-remediation, Step 6 warn-and-skip and per-hit skip), but there is no true retry loop and the wiki step's pseudocode slightly muddies execution sequence.	4 / 5
Progressive Disclosure	Clear sectioned structure with one-level-deep references clearly signaled via markdown links to shared-references; the main gap is the inlined, duplicated resolution logic that arguably belongs in the referenced contract.	4 / 5
	Total	16 / 20 Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that answers both what and when with concrete quoted triggers and explicit differentiation from sibling academic-search skills. The only gap is action specificity falling just short of fully comprehensive.

Dimension	Reasoning	Score
Specificity	Names the Exa domain and lists several concrete actions — 'web search', 'content extraction', 'find similar pages' — but stops short of the comprehensive multi-action list of a 5.	4 / 5
Completeness	Clearly states what it does ('AI-powered web search via Exa with content extraction') and explicitly answers when with concrete trigger phrases and the 'beyond academic databases' condition.	5 / 5
Trigger Term Quality	Quotes natural trigger phrases users would say ('exa search', 'web search with content', 'find similar pages') with good coverage, though a few common synonyms like 'search the web' are missing.	4 / 5
Distinctiveness Conflict Risk	Carves a clear niche as broad web search and explicitly differentiates against arXiv and Semantic Scholar, giving minimal conflict risk with related academic skills.	5 / 5
	Total	18 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 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
relative_links	Relative link issues: 4 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/exa-search/SKILL.md
Commit: f4f20f9

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.