pr-slicer

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses git and gh (GitHub CLI) to automate branch management and PR creation.
  • Evidence: Workflow steps specify commands like git switch, git checkout, git cherry-pick, git rebase, git push, and gh pr create.
  • Evidence: Step 9 ("Validate each PR accurately") instructs the agent to run the repository's local validation scripts (e.g., tests), which involves executing code found within the user's workspace.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes repository content (diffs, commits) which presents an attack surface for indirect prompt injection.
  • Ingestion points: File SKILL.md (Steps 2 and 9) requires the agent to read branch diffs and repository files.
  • Boundary markers: Absent; there are no instructions to treat repository content as untrusted data or use specific delimiters to isolate it from instructions.
  • Capability inventory: File SKILL.md (Steps 7, 8, 9, 11, 14) grants capabilities for git operations, GitHub interaction, and execution of local validation scripts.
  • Sanitization: Absent; the skill does not specify any filtering or escaping of ingested data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 03:55 PM
Security Audit — agent-trust-hub — pr-slicer