openspec-explore
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates external data from the local codebase and configuration files, which serves as a potential surface for indirect prompt injection.
- Ingestion points: Reads project background, tech stack conventions, and constraints from
openspec/config.yamland existing design artifacts. - Boundary markers: The instructions do not specify the use of delimiters or explicit boundary markers to isolate ingested file content from agent instructions.
- Capability inventory: The skill executes specific commands via the
openspecCLI to manage design artifacts and status. - Sanitization: No content validation or sanitization procedures are described for the data read from the local environment.
Audit Metadata