skills/arcblock/agent-skills/pr-sweep/Gen Agent Trust Hub

pr-sweep

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes git and gh CLI tools to synchronize the repository state and manage Pull Requests. Automated commands include git fetch, git reset --hard, git clean -fd, gh pr list, and gh pr merge.
  • [COMMAND_EXECUTION]: The agent executes repository-specific scripts defined in a repo-profile.md file, such as <verification_entry>, <pre_merge_entry>, and <merge_gate_entry>, which are used to validate PRs before merging.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its ingestion of untrusted data from GitHub PR titles, bodies, and comments (identified in Step 1 and 1.5). The skill lacks explicit sanitization or strict boundary markers for this external content, which could allow a malicious actor to influence agent behavior given its extensive capabilities, such as shell command execution and PR modification.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 10:26 PM
Security Audit — agent-trust-hub — pr-sweep