prune-branches
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (tool output poisoning) because it ingests branch names from the local repository and interpolates them directly into shell commands. Although the risk is mitigated by manual user confirmation, the skill lacks explicit sanitization or boundary markers for this untrusted data.
- Ingestion points: The skill reads branch names using
git branch --merged(SKILL.md). - Boundary markers: Absent; branch names are interpolated directly into command strings.
- Capability inventory: The skill can execute branch deletion (
git branch -d) and remote push operations (git push --delete) (SKILL.md). - Sanitization: Absent; the skill relies on the user to review the list of branches in the
AskUserQuestionstep before execution.
Audit Metadata