blueprint
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted local project data, which could contain malicious instructions designed to influence the agent's behavior during the planning phase.
- Ingestion points: The Research phase reads project structure, existing plan files, and local memory files from the workspace (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted project content from the skill's own operational instructions.
- Capability inventory: The skill utilizes the agent's ability to read and write files within the repository and interact with the local git and GitHub CLI environments.
- Sanitization: The instructions do not specify any validation, filtering, or escaping techniques for data retrieved from external project files.
- [SAFE]: The installation section includes shell commands (e.g., git fetch, test -f), but these are provided as documentation for the user's manual execution and are not invoked automatically by the skill's instructions.
- [SAFE]: The skill contains no executable code, build steps, or dynamic runtime components, relying entirely on the native Markdown capabilities of the host agent.
Audit Metadata