openserp-search
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The script utilizes the
uvxtool to download and execute theopenserpPython package. This package is an external dependency retrieved from the public PyPI registry. - [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it fetches and processes untrusted data from the internet (search titles, snippets, and page content).
- Ingestion points: External data enters the agent's context through search results processed in
scripts/search. - Boundary markers: The search results are output directly to the agent's context without using delimiters or instructions to treat the content as untrusted.
- Capability inventory: The agent utilizing this skill has access to the
shelltool, which could be targeted by instructions hidden in search results. - Sanitization: There is no evidence of content sanitization or filtering for the retrieved web snippets.
Audit Metadata