unifuncs-search
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the web.
- Ingestion points: Web search results, including page names, snippets, and summaries, are retrieved from the vendor API in
scripts/search.pyand output to the agent context. - Boundary markers: No specific delimiters or boundary markers are used to separate search results from instructions in the formatted output.
- Capability inventory: The skill is granted
Bash(python*:*)permissions to execute the search script. - Sanitization: The search results are not sanitized or validated before being presented to the agent, which could allow malicious instructions in search snippets to influence agent behavior.
Audit Metadata