solution-feasibility-study
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from various external sources, which could contain malicious instructions designed to influence the agent's research conclusions or recommendations. 1. Ingestion points: The references/research-playbook.md file instructs the agent to use WebFetch on commercial competitor websites and GitHub repository pages. 2. Boundary markers: Absent. 3. Capability inventory: The skill utilizes WebSearch, WebFetch, file system writes for report generation, and shell command execution via the gh CLI suggestion. 4. Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill incorporates the use of shell commands based on external data found during research. 1. Evidence: SKILL.md and references/report-template.md include the command
gh repo fork <org>/<repo> --cloneas a suggested next step for the user. The parameters are derived from untrusted web research.
Audit Metadata