review-all-gdds
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits a significant attack surface for indirect prompt injection due to its core function of reading and processing numerous local project files.
- Ingestion points: The skill reads AGENTS.md, CLAUDE.md, all markdown files in design/gdd/, the entities.yaml registry, and other documentation files.
- Boundary markers: There are no specific instructions or delimiters used to ensure the agent ignores instructions that may be embedded within these documents.
- Capability inventory: The agent has access to powerful tools, including Bash (used for git log), Write (used for report generation and state updates), and Task (used for spawning sub-agents).
- Sanitization: The skill does not perform any sanitization or validation of the ingested content before processing it or passing it to sub-agents.
- [DYNAMIC_EXECUTION]: The workflow involves spawning parallel Task agents and constructing their prompts using the full content of the registry and loaded GDD documents. This dynamic prompt construction allows untrusted content to influence the instructions given to the sub-agents.
Audit Metadata