cut-release
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands including git for repository management (tagging, diffing, logging) and the GitHub CLI (gh) for release and issue tracking updates. Evidence is present in SKILL.md sections a, b, d, e, g, h, and i.\n- [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by reading and executing commands defined in external project files, though this is for the intended purpose of project builds.\n
- Ingestion points: The skill reads build, verify, and smoke commands from docs/agents/project.md.\n
- Boundary markers: Absent; the skill relies on the formatting of the project documentation to identify commands.\n
- Capability inventory: The skill has the capability to execute shell commands as specified in the project's docs/agents/project.md file.\n
- Sanitization: Absent; there is no validation of the commands fetched from the documentation before they are executed in the shell.\n- [DYNAMIC_EXECUTION]: The skill dynamically loads and executes the build and release sequence based on the content of the repository's documentation at runtime.\n
- Evidence: Section e of SKILL.md requires the agent to execute release steps found in docs/agents/project.md exactly as written.
Audit Metadata