openspec-onboard
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several local commands including
openspec status,git log,openspec new change, andopenspec archiveto facilitate the onboarding workflow. These commands are aligned with the skill's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the local codebase by scanning for
TODO,FIXME, and other patterns to suggest improvements. This creates a surface where external data (code comments) can influence the agent's generated output and actions. - Ingestion points: Codebase files and git history scanned in Phase 2.
- Boundary markers: The instructions do not define delimiters or specific safety instructions for handling ingested code content.
- Capability inventory: The skill can create directories, write files, and execute the
openspecCLI tool. - Sanitization: There is no explicit sanitization or filtering of the codebase snippets before they are processed by the agent.
Audit Metadata