create-rc-tag
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple bash commands and Git operations to manage release workflows. This includes fetching from remotes, switching branches, creating signed tags, and pushing to the 'upstream' remote.
- [DYNAMIC_EXECUTION]: The skill uses
python -cto execute a one-line Python script that parsesCargo.tomlusingtomllib. The Python code is static and hardcoded within the skill instructions, which minimizes the risk of executing arbitrary data. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied input to determine the release version, creating a potential surface for injection.
- Ingestion points: The variable
RELEASE_VERSION_INPUTis populated directly from user input inSKILL.md. - Boundary markers: The skill does not use specific delimiters or instructions to the agent to ignore instructions embedded in the version string.
- Capability inventory: The skill has access to shell execution (
bash), Git operations, and Python execution. - Sanitization: The skill implements strict sanitization via regex validation (
^[0-9]+\.[0-9]+$and^[0-9]+\.[0-9]+\.[0-9]+$), which ensures the input only contains numeric version components, effectively preventing command injection or complex prompt injection via this vector.
Audit Metadata