openspec-apply-change
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill interacts with the local
openspecCLI to list changes, check status, and retrieve implementation instructions. These interactions are scoped to the tool's intended purpose and do not involve unauthorized network access or sensitive file exposure. - [SAFE]: Implementation logic includes multiple guardrails, such as requiring the agent to pause for clarification if tasks are ambiguous, keeping code changes minimal and focused, and reporting blockers rather than proceeding blindly.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external context files (proposals, specs, design documents) and CLI-generated dynamic instructions to drive the implementation loop. While this creates a surface for indirect instructions, the skill mitigates this by instructing the agent to focus strictly on task-based code changes and to pause whenever requirements are unclear or implementation issues arise.
Audit Metadata