scarce-resource

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub Pull Request labels, bodies, and commit messages to determine if a 'waiver' exists to bypass scheduling constraints.
  • Ingestion points: Fetches data via gh pr list --label waits:<lane> and inspects the PR body and head commit message for specific markers.
  • Boundary markers: There are no instructions to use delimiters or ignore potentially malicious instructions embedded within the PR or commit content.
  • Capability inventory: The skill can dispatch CI workflows (orch-dispatch), manipulate background processes (orch-wakeat), and create systemd user timers.
  • Sanitization: No sanitization, escaping, or validation of the PR body or commit messages is mentioned before the agent evaluates them for waivers.
  • [COMMAND_EXECUTION]: The skill uses several external scripts (orch-wakeat, orch-dispatch, orch-idlewait, orch-runwait, orch-conflicts) and the GitHub CLI (gh) to perform resource orchestration. These scripts are expected to be available on the system's PATH.
  • [DATA_EXPOSURE]: The skill explicitly accesses configuration files located at ~/.config/orchestrate/<repo>.env, which may contain sensitive repository-specific environment variables or credentials required for the orchestration scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 09:10 AM