openspec-proposal
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feature descriptions and includes them in generated markdown files. This is a potential surface for indirect prompt injection if those files are later interpreted as instructions by an AI agent.\n- Ingestion points: The
<feature description>parameter provided by the user in the command/openspec-proposal.\n- Boundary markers: No delimiters are used to separate user input from template structure in the generated files.\n- Capability inventory: The skill usesRead,Write,Glob,Grep, andBash(formkdir) to manage project files.\n- Sanitization: Path traversal is mitigated by the instruction to convert user input to 'kebab-case' for directory names.
Audit Metadata