senior-fullstack
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to analyze untrusted project data and execute project-defined scripts, which creates a surface for indirect prompt injection attacks where malicious instructions in project files could influence the agent's behavior.
- Ingestion points: Project dependency manifests (e.g., package.json, requirements.txt), lockfiles, entrypoints, and build configuration files as described in SKILL.md and references/tech_stack_guide.md.
- Boundary markers: No specific delimiters or instructions to ignore embedded natural language instructions within the project files are provided.
- Capability inventory: The skill workflow directs the agent to execute "project-owned checks" and build commands (e.g., tests, linting, type-checking) which involves running arbitrary scripts defined in untrusted project configurations.
- Sanitization: No sanitization, validation, or integrity checking of the external project scripts or manifests is mentioned before processing or execution.
Audit Metadata