create-branch
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard Git commands (
git diff,git log,git checkout) to gather context and perform branch operations. These are appropriate for the skill's stated purpose of managing a development workflow. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Git commit messages and diffs to derive branch names. However, it implements a robust sanitization process via
validate.sh, which enforces a strict allowlist of characters (lowercase alphanumeric and hyphens) and predefined types (e.g., feat, fix). This prevents malicious content in Git logs from influencing the command line execution. - Ingestion points: Git diffs, status, and log messages are read into the agent's context (SKILL.md, Step 1).
- Boundary markers: None explicitly defined in the instructions, but naming conventions serve as implicit boundaries.
- Capability inventory:
git checkout -b <name>is the primary action (SKILL.md, Step 7). - Sanitization: High. The
validate.shscript strictly validates the proposed branch name using regex before the command is executed. - [EXTERNAL_DOWNLOADS]: No external dependencies or remote scripts are downloaded; the skill relies entirely on locally provided files and standard system tools.
Audit Metadata