framework-docs-researcher
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill directs the agent to execute shell commands such as 'bundle show' to determine the installation paths of libraries.
- [DATA_EXFILTRATION] (LOW): The agent is instructed to read local project files like 'Gemfile.lock' and explore the source code of installed gems to extract version and implementation details. This constitutes data exposure of local project components.
- [PROMPT_INJECTION] (LOW): The agent processes information from untrusted external sources, including web search results and GitHub discussions, which presents a surface for indirect prompt injection. Evidence: 1. Ingestion points: Web search results and GitHub repository data (issues, PRs, and source). 2. Boundary markers: Absent. 3. Capability inventory: Local command execution via 'bundle show' and filesystem read access. 4. Sanitization: None mentioned in the instructions.
Audit Metadata