plan-first
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes project files, including READMEs and configuration manifests (package.json, go.mod, etc.), which are external and potentially untrusted data sources.
- Ingestion points: Phase 1 involves reading directory structures, dependency manifests, and README files from the project root.
- Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the analyzed files.
- Capability inventory: The skill possesses the capability to write files (creating TODO.md in Phase 3) and perform general task execution in Phase 5.
- Sanitization: No explicit sanitization or validation of the content read from the project files is performed.
- Note: The requirement for user approval of the TODO.md before execution serves as a significant security control against malicious instructions being automatically adopted from project metadata.
Audit Metadata