orchestration
Run work in separate Orca worker agents and get their results back. Use when the user asks for Orca workers or supervision (monitor, task DAGs, ask/reply, decision gates), or when the work needs something your own subagents cannot give: a different agent (such as Codex from Claude) or its own Orca workspace. "Have Codex (or Claude) review this and report back" means an Orca worker of that agent, even when you have your own model subagents. Also use when you are a dispatched worker (a prompt with Task and Dispatch IDs: ask, worker_done). Otherwise prefer your own subagents, including for long or parallel work. Use `orca-cli` to hand work off with nothing expected back ("hand off", "handover", "give this to another agent", "another worktree"), and for terminal control, shell commands, worktree management, and reading or waiting on terminals.
- Created
- 8 months ago
- Last commit
- 2 days ago
- Latest release
- v1.4.2583 days ago
Security Findings
Unpinned dependencies
SuspiciousLineage1 unpinned: @orca/process-host
Flagged by Gen Agent Trust Hub on skills.sh (warn · medium)2×
SuspiciousLineagePRIVILEGE_ESCALATION; COMMAND_EXECUTION; INDIRECT_PROMPT_INJECTION; DYNAMIC_EXECUTION — [PRIVILEGE_ESCALATION]: The instructions explicitly direct the agent to re-run commands with escalated permissions if a runtime_access_denied error is returned, which is an attempt to bypass environment or sandbox security constraints. [COMMAND_EXECUTION]: The skill is centered around executing shell commands via the Orca CLI suite (orca, orca-ide, orca-dev) to manage workers and workspaces. [INDIRECT_PROMPT_INJECTION]: The skill fetches its primary operational guide at runtime by executing ORCA skills get orchestration and incorporating the output into its context. This presents a vulnerabil…
W013 Attempt to modify system services in skill instructions. — Attempt to modify system services in skill instructions detected (low risk: 0.35). The skill mainly orchestrates Orca CLI commands and does not ask for sudo, system file edits, user creation, or service changes, but it tells the agent to re-run blocked commands with escalated permissions, which is a mild sandbox-bypass concern.
Versions
11- Scanned Nov 15, 2026, 08:22 AMLow
- Scanned Nov 8, 2026, 08:15 AMLow
- Scanned Nov 1, 2026, 06:35 PMLow
- Scanned Oct 31, 2026, 01:27 PMLow
- Scanned Oct 29, 2026, 11:56 PMLow
- Scanned Oct 25, 2026, 01:17 PMLow
- Scanned Oct 9, 2026, 02:50 PMLow
- Scanned Aug 17, 2026, 02:54 PMLow
- Scanned Aug 17, 2026, 02:54 PMLow
- Scanned Aug 17, 2026, 02:54 PMLow
- Scanned May 6, 2026, 06:38 PMLow