openspec-archive-change
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands such as
mkdirandmvto create archive directories and relocate project folders. These operations are limited to the project's own directory structure (openspec/changes/). - [PROMPT_INJECTION]: The skill processes data from local task files and CLI outputs, representing an indirect prompt injection surface.
- Ingestion points:
tasks.md,openspec list --json,openspec status --json. - Boundary markers: The skill utilizes structured JSON parsing for CLI command outputs, which reduces the risk of parsing-based injection.
- Capability inventory: File system modifications (
mkdir,mv) and subagent invocation for synchronization tasks. - Sanitization: No explicit sanitization of markdown task lists is mentioned, but risk is minimized as the data is used for status counting rather than command generation.
Audit Metadata