Web SearchSAFE
A framework-agnostic, git-native standard for defining AI agents
Overview
A framework-agnostic, git-native standard for defining AI agents
799d6d0ad18cOBSERVED · 2026-10-07What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: web-search description: Advanced web search using Tavily API for current information retrieval allowed-tools: tavily-web-search --- # Web Search Search the web for current information on any topic using the Tavily search API. ## When to Use - Searching for recent news, articles, blog posts, and web content - Finding current statistics, data, and factual information - Retrieving information that may not be in academic papers or the knowledge base - Verifying claims with multiple web sources ## How to Use 1. Formulate a specific, targeted search query (max 400 characters) 2. Call `tavily_web_search` with the query 3. Review returned results for relevance and quality 4. Extract key facts and note the source URL for citation ## Result Format Results are returned in Document XML format: ```xml <Document href="https://example.com/article" title="Article Title"> Content excerpt from the page... </Document> ``` ## Constraints - Maximum 5 results returned per query - Query will be truncated to 400 characters - Search mode: advanced (includes content extraction) - Each call counts toward the researcher's 8-call limit per task
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
799d6d0ad18cfull audit observations/trust-audit/skill/open-gitagent__web-search.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 799d6d0ad18c | SAFE | B | 89 | first audit |
Questions
What does the Web Search skill do?
A framework-agnostic, git-native standard for defining AI agents
Is Web Search safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Web Search access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (799d6d0ad18c), read on 2026-10-07. The repository is watched, and a new audit runs when it changes — this is the first audit.