CtrlK
BlogDocsLog inGet started
Tessl Logo

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).

63

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/exa-search/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

62%

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

The skill body is well-structured with a clear, validated multi-step workflow and mostly executable guidance. It loses points for the inline pseudocode mixed into a bash fence in Step 6, duplicated resolution boilerplate that could be referenced out, and the absence of any local bundle files to support progressive disclosure.

Suggestions

Replace the pseudocode control-flow inside the Step 6 bash fence with either runnable bash or clearly separated non-code prose, so the code block is fully executable.

Factor the duplicated helper-resolution cascade (EXA_FETCHER and WIKI_SCRIPT) into the referenced integration-contract so the SKILL.md body references it once instead of inlining it twice.

Move the full override/flag reference list (Step 1) into a local references/ file and keep only the most common examples inline, improving token efficiency and progressive disclosure.

DimensionReasoningScore

Conciseness

The body is mostly operational and efficient, but contains tightening opportunities such as the verbatim-repeated bash resolution cascade (EXA_FETCHER then WIKI_SCRIPT) and minor padding like "Default to highlights content mode for a good balance of speed and context".

2 / 3

Actionability

Core search commands (search/find-similar/get-contents with --category, --start-date, --content text --max-chars) are concrete and executable, but Step 6 embeds non-executable pseudocode control-flow ("if URL matches arxiv.org/abs/<id>:" / "for each research-paper hit in results:") inside a bash code fence, mixing pseudocode with executable code.

2 / 3

Workflow Clarity

A clear six-step sequence (Parse, Locate, Execute, Present, Follow-up, Wiki) with explicit validation/feedback loops: Step 2 fails with explicit remediation when the helper is unresolved, and Step 6 uses a warn-and-skip (Variant B) recovery when the wiki helper is missing.

3 / 3

Progressive Disclosure

Sections are well-organized and shared-references (integration-contract.md §2, wiki-helper-resolution.md) are clearly signaled at one level deep, but no local bundle files exist, the referenced files live outside this skill bundle, and the lengthy inline bash resolution cascade is content that could be externalized rather than duplicated.

2 / 3

Total

9

/

12

Passed

Description

90%

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 description: third-person, explicit "Use when" triggers with natural user phrasing, and a clear niche against sibling academic-search skills. Its only weakness is specificity — it states the domain and content extraction but does not list the multiple concrete operations the skill supports.

DimensionReasoningScore

Specificity

"AI-powered web search via Exa with content extraction" names the domain and one concrete capability (content extraction) but does not enumerate the full set of actions (search, find-similar, get-contents, category/date/domain filtering), so it is not comprehensive.

2 / 3

Completeness

Explicitly answers what ("AI-powered web search via Exa with content extraction") and when (an explicit "Use when..." clause with concrete triggers), satisfying both halves with explicit trigger guidance.

3 / 3

Trigger Term Quality

Quotes natural user phrases — "exa search", "web search with content", "find similar pages" — plus "broad web results beyond academic databases (arXiv, Semantic Scholar)", giving good coverage of terms a user would actually say.

3 / 3

Distinctiveness Conflict Risk

The carve-out "beyond academic databases (arXiv, Semantic Scholar)" and the unique trigger "exa search" give it a clear niche unlikely to conflict with the sibling academic search skills.

3 / 3

Total

11

/

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.

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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
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.