auto-release-manager
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill exhibits a surface for Indirect Prompt Injection through its changelog generation feature.
- Ingestion points: Git commit history (untrusted data) is read by
scripts/changelog_generator.pyin Step 4. - Boundary markers: The instructions do not define delimiters or warnings for the agent to ignore instructions embedded within commit messages.
- Capability inventory: The skill has access to
Bash(command execution),Write(file modification), and can performgit pushoperations. - Sanitization: There is no evidence of sanitization or filtering applied to the commit messages before they are processed by the generator or presented to the agent.
- [COMMAND_EXECUTION] (SAFE): The skill executes several local scripts (
scripts/detect_project.py,scripts/update_version.py,scripts/sync_unity_version.py,scripts/git_operations.py) and uses theghCLI. These operations are essential for the primary purpose of version and release management and do not appear to involve untrusted remote sources or dynamic code generation.
Audit Metadata