git-detect-breaking-changes
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes source code and git history from the local repository and incorporates this data into the agent context.\n
- Ingestion points: Repository files and git metadata are read using
Read,Glob,Grep, andgit show(SKILL.md).\n - Boundary markers: The skill lacks explicit markers or instructions to delimit or ignore instructions that may be embedded in the analyzed data.\n
- Capability inventory: The skill possesses file read access and command execution capabilities via
Bash,Read,Glob, andGrepacross the workspace.\n - Sanitization: The skill does not perform sanitization of the external content before interpolation into the agent's prompt.\n- [COMMAND_EXECUTION]: The skill performs shell command execution using the
Bashtool to rungitsubcommands (log,show,diff,rev-parse). These commands utilize user-provided input ($ARGUMENTS), which represents a potential command injection surface if the underlying platform does not provide strict input validation.
Audit Metadata