gigio-project-setup
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for project documentation and context management. It provides a framework for the agent to recover, write, and audit project-related information such as goals, constraints, and resource mappings without employing active code execution or external network communication.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to process existing project documentation and instructions, which introduces an indirect injection surface.
- Ingestion points: The agent is instructed to inspect 'existing intent document, project instructions, relevant decisions, current results, and resource locations' (SKILL.md).
- Boundary markers: The skill does not provide specific delimiters or warnings to ignore instructions embedded within the ingested project data.
- Capability inventory: The agent's capabilities in this context are limited to modifying and creating project records (PROJECT.md) and instruction entry points (AGENTS.md, CLAUDE.md).
- Sanitization: No sanitization or filtering protocols are defined for the ingested content. Since the resulting actions are limited to updating documentation and do not involve high-risk capabilities, the finding is categorized as safe.
Audit Metadata