onboarding
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from repository-hosted files including
README,Makefile,docker-compose.yml, and CI configurations to determine setup procedures. This creates an attack surface where a compromised repository could provide malicious setup instructions. - Ingestion points:
repo-map.md,docker-compose.yml,Makefile, CI config,READMEreferenced in Step 2 and Step 3. - Capability inventory: Shell command execution via
uv,poetry,pip, andmanage.pyas described in Step 2. - Boundary markers: Absent; the agent is instructed to derive steps directly from the files.
- Sanitization: Absent; the skill relies on the developer to verify the derived steps.
- [COMMAND_EXECUTION]: The skill directs the execution of shell commands for package management (
uv sync,poetry install,pip install), environment setup, and database operations. These are consistent with the skill's primary purpose of developer onboarding and use standard industry tools.
Audit Metadata