openspec-journal
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a utility for project-local logging and interaction tracking. It does not perform network operations, request elevated privileges, or attempt to persist itself beyond providing documentation for manual hook installation.
- [PROMPT_INJECTION]: The instructions in SKILL.md define a behavioral protocol for the agent to log 'turn.start' and 'turn.end' events. These are functional requirements for the skill's logging feature and do not attempt to override the agent's safety guidelines or core instructions.
- [COMMAND_EXECUTION]: The skill utilizes a Python script and a bash script for its core functionality. These scripts execute standard system commands (like 'git' and 'stat') and local Python code without processing untrusted external input in a way that leads to command injection.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided summaries and workspace file content. It mitigates potential injection risks by enforcing a 200-character limit on summary fields, stripping newlines, and using standard JSON encoding to store the data.
Audit Metadata