openspec-verify-change
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several commands via the
openspecCLI to gather project metadata and artifact paths for analysis. - Evidence:
openspec list --json,openspec status --change "<name>" --json, andopenspec instructions apply --change "<name>" --json. - [EXTERNAL_DOWNLOADS]: The skill requires the external
openspecCLI tool to be present on the system to function. This dependency is declared in the skill metadata. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from local project files, including task lists and requirement specifications.
- Ingestion points: Reads files found in the
contextFileslist returned by the CLI and searches the local codebase (SKILL.md). - Boundary markers: None identified.
- Capability inventory: Limited to read-only analysis and executing the
openspecCLI; no arbitrary file write or external network operations are specified. - Sanitization: Not present; the agent parses requirements and scenarios directly from markdown artifacts.
Audit Metadata