openspec-propose
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes the
openspecCLI tool to perform various project management tasks, including change initialization, status tracking, and instruction retrieval.- [DYNAMIC_EXECUTION]: The workflow allows for the invocation of additional commands or skills if specified within the instruction data returned by the OpenSpec CLI during the artifact creation process.- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific context and instructions provided by theopenspecCLI and existing artifact files. Ingestion points: Data retrieved viaopenspec instructionsand existing artifact files on disk. Boundary markers: The instructions include a strict 'Planning boundary' that restricts the agent's actions to planning artifacts and forbids implementation or project code editing. Capability inventory: The skill utilizesBash(openspec:*)for command execution and has the capability to write files to the local filesystem. Sanitization: The skill contains explicit instructions to treat provided context and rules as constraints for generating content rather than including them directly in the final output.
Audit Metadata