herdr

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document primarily describes legitimate CLI orchestration but includes explicit, repeated instructions to disable per-command permission prompts (e.g., --yolo/--force/--dangerously-skip-permissions), which intentionally bypass safety controls and can enable unauthorized remote actions by unattended agents.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). SKILL.md’s required runtime workflow uses herdr pane read <pane-id> --source recent-unwrapped --lines 120 to ingest the target pane’s recent output (which can include outsider-authored free text from other agents/users) into the controlling agent’s LLM context.