openspec-verify-change

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious behavior or security violations were identified. The skill's operations are consistent with its documented purpose.
  • [COMMAND_EXECUTION]: The skill executes commands using the openspec CLI to fetch change lists and statuses. These operations are restricted to the local environment and use variables derived from user-selected change names.
  • [DATA_EXFILTRATION]: The skill accesses local files such as tasks.md, design.md, and specification files to compare them against the implementation. It does not contain any network-enabled tools or instructions to transmit this data externally.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external project artifacts. It analyzes these files for specific structural elements like checkboxes and requirement headers to build a report. It does not treat the content of these files as instructions to the agent, which effectively mitigates the risk of indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 12:09 AM