tavily-extract — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited tavily-extract (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Extract clean markdown/text content from one or more URLs. In this repo, "opened" means extracted with this skill (or fetched with a built-in web tool) — only opened URLs are citable as evidence.
Check tvly --status. If tvly is not on PATH:
pip install -q tavily-cli || uv tool install tavily-cliAuth comes from the TAVILY_API_KEY environment variable (already set in this environment). Never print it. If it is unset, report that and fall back to the built-in web tools.
# Single URL
tvly extract "https://example.com/article" --json
# Batch (max 20 per call)
tvly extract "https://arxiv.org/abs/XXXX.XXXXX" "https://github.com/org/repo" --json
# Long pages: only the relevant chunks
tvly extract "https://example.com/docs" --query "release date changelog" --chunks-per-source 3 --json
# JS-heavy pages
tvly extract "https://app.example.com" --extract-depth advanced --json| Option | Description |
|---|---|
--query + --chunks-per-source | Return only chunks relevant to a query (1–5 per URL) |
--extract-depth | basic (default, try first) or advanced (JS-rendered pages) |
--format | markdown (default) or text |
--timeout | 1–60 seconds for slow pages |
--json / -o FILE | Structured output / save to file |
? and &).basic first; escalate to advanced only if content is missing.metadata with curl -sL 'https://export.arxiv.org/api/query?id_list=...' (see radar-source-verify).
--include-raw-content, skip the extract step.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.