blueprint
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests user objectives to generate plans, creating a surface for indirect prompt injection where malicious instructions in the input could influence the generated Markdown content.
- Ingestion points: User-supplied objective text provided to the slash command.
- Boundary markers: No explicit delimiters are used to separate user input from system instructions.
- Capability inventory: The skill reads directory structures and writes Markdown files to a 'plans/' folder.
- Sanitization: No input sanitization or filtering is described in the skill workflow.
- [SAFE]: The skill is composed of pure Markdown instructions with no executable scripts, build steps, or remote dependencies, significantly limiting the potential impact of any injection.
- [SAFE]: Commands referenced (git, gh) are used for environment status checks and local repository management tasks within a standard developer workflow.
Audit Metadata