CtrlK
BlogDocsLog inGet started
Tessl Logo

web-search-tavily

AI-powered web search, crawling, extraction, and deep research

52

Quality

58%

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/research-tools/capabilities/web-search-tavily/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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.

The body is well-sectioned and gives mostly executable curl guidance, but it inlines a large API parameter reference that should live in a separate file and ships two broken curl examples for its primary capabilities. Fixing the malformed examples and splitting out the reference would materially raise the score.

Suggestions

Fix the Tavily Search and Tavily Crawl curl blocks so all JSON fields sit inside the `-d` quoted body string, matching the working Extract/Map/Research examples.

Move the per-endpoint parameter tables into a references/ file (e.g. ENDPOINTS.md) and link to it from each section, keeping only one worked example inline.

Add an explicit research-task workflow (create task → poll /research/{request_id} until complete → read report) with a validation/check-status checkpoint.

DimensionReasoningScore

Conciseness

Mostly efficient API-oriented prose, but ~120 lines of per-endpoint parameter reference are inlined and the chunks_per_source description is repeated verbatim across Search, Extract, and Crawl sections, adding avoidable tokens.

3 / 5

Actionability

Concrete curl examples are provided per endpoint, but the Tavily Search and Tavily Crawl examples are malformed — the JSON body fields dangle outside the `-d` quoted string — so two primary examples are not copy-paste executable.

3 / 5

Workflow Clarity

Setup steps are sequenced and the async research create→poll-status pattern is implied, but there are no explicit validation checkpoints (e.g. verifying credentials, polling guidance for research tasks) and the multi-step research flow is not spelled out.

3 / 5

Progressive Disclosure

The body is sectioned by capability, but the bulk of API parameter reference that belongs in a separate file is inlined into SKILL.md; no local reference bundle files exist, though the "Discover More" section points to live discovery endpoints.

3 / 5

Total

12

/

20

Passed

Description

66%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.

The description is concise and names concrete capabilities with good trigger keywords, but it omits any explicit "Use when…" trigger guidance, leaving Claude without clear activation cues. Adding a when-to-use clause would lift completeness and distinctiveness.

Suggestions

Append a trigger clause, e.g. ".Use when the user needs to search the web, crawl or extract content from URLs, or run deep research on a topic."

Add natural synonyms such as "scrape", "index a documentation site", or "find online sources" to broaden trigger coverage.

Drop the filler "AI-powered" prefix to keep every token earning its place.

DimensionReasoningScore

Specificity

Names the domain and several concrete actions ("web search, crawling, extraction, and deep research"), giving broad coverage of the API's capabilities, though "deep research" is somewhat abstract and "AI-powered" is light fluff.

4 / 5

Completeness

The description clearly states what the skill does but lacks any "Use when…" clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Includes natural terms users say ("web search", "crawling", "extraction", "deep research") with good coverage, but misses common synonyms like "scrape", "index a site", or "find sources online".

4 / 5

Distinctiveness Conflict Risk

The combination of search/crawl/extract/research carves out a fairly distinct niche with only minor overlap risk against a generic web-scraping skill, though the absence of trigger phrases slightly weakens distinctiveness.

4 / 5

Total

15

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
gooseworks-ai/goose-skills
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.