openspec-bulk-archive-change
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes the
openspecCLI utility for listing and checking the status of changes. It also uses shell commands likemkdir -pandmvto restructure the local file system during the archiving process. - [PROMPT_INJECTION]: Potential for Indirect Prompt Injection exists as the skill processes natural language data from the project's codebase to resolve implementation conflicts.
- Ingestion points: Processes
openspec/changes/<name>/tasks.md, directory listings inopenspec/changes/<name>/specs/, and performs general codebase searches for implementation evidence. - Boundary markers: Absent. The instructions do not define delimiters or provide specific guidance to the agent to disregard instructions found within the analyzed files.
- Capability inventory: Includes file system modification (
mv,mkdir), execution of theopenspecCLI, and reading arbitrary files within the repository. - Sanitization: Absent. There is no evidence of validation or filtering of content retrieved from the codebase before it is used to drive the agent's logical conclusions regarding conflict resolution.
Audit Metadata