Orca orchestration: supervise multi-agent workers
Coordinate supervised Orca workers: threaded messages, blocking ask/reply, task DAGs, decision gates and coordinator loops
Curated by Skill Harbor — @stablyai's orchestration skill for the Orca CLI: a structured way to run supervised multi-agent work — threaded messages between agents, blocking ask/reply flows, task dispatch with worker_done/escalation waits, task DAGs, decision gates, coordinator loops, and decomposing work across agents and worktrees. By design this file is a discovery stub: the full version-matched orchestration reference is served live by the `orca` binary itself (`ORCA skills get orchestration`) so the guide can never drift from the binary that actually runs your commands. Honest caveats: requires a working Orca runtime (the skill helps you resolve the right executable — `orca-dev`, `orca-ide` or `orca` — and warns against Linux's GNOME Orca screen reader of the same name); MIT-licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Quick answer
Orca Orchestration is a skill for running supervised multi-agent work through the Orca CLI: threaded messages between agents, blocking ask/reply, task dispatch with decision gates and coordinator loops. By design the file is a discovery stub — the full, version-matched orchestration reference is served live by the orca binary itself, so the guide can never drift from the tool that runs your commands. It requires a working Orca runtime.
At a glance
| What it is | A structured way to coordinate supervised Orca workers: threaded messages, blocking ask/reply flows, task dispatch with worker_done and escalation waits, task DAGs, decision gates, coordinator loops, and decomposing work across agents and worktrees. |
|---|---|
| Who it's for | Developers already running the Orca CLI who coordinate several agent workers and want supervision structure — gates and escalations — instead of free-running agents. |
| What you give it | A working Orca runtime on your machine and the work to decompose. The skill helps resolve the right executable — orca-dev, orca-ide or orca — and warns against Linux's GNOME Orca screen reader, which shares the name. |
| What you get | A supervised orchestration pattern over your Orca workers: tasks dispatched through a DAG, questions and answers blocking where they should, decision gates and escalation waits, all following the live reference served by the orca binary itself. |
| Works with | The Orca CLI runtime — this skill does not replace it. The orchestration reference comes from the orca binary (ORCA skills get orchestration), so it always matches your installed version; install the skill from this listing with the copy-paste install prompt. |
| Cost | Free |
| Time | A few minutes to install the skill, on top of having Orca running; each orchestration then lasts as long as the multi-agent work it supervises. |
Example — dispatching supervised work
An example of a typical use, using only the coordination patterns the skill documents — not a reported result.
- You break a job into a task DAG and dispatch the pieces to supervised Orca workers, each in its own worktree.
- A worker hits a decision it should not make alone; the blocking ask/reply flow escalates the question and waits for the answer before continuing.
- When a worker finishes, its worker_done signal lets the coordinator loop release the next dependent task in the DAG.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox and failed · Nov 18, 2026 · Needs pnpm, npm cannot install it📌 Pinned version: commit c9c6a00Step 0 — before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: a working Orca runtime (the orca CLI installed and running); no accounts or API keys required Install "Orca orchestration: supervise multi-agent workers" for me. It teaches structured multi-agent coordination with the Orca CLI — threaded messages, blocking ask/reply, task dispatch with worker_done/escalation waits, task DAGs, decision gates, coordinator loops, and work decomposition across agents — by @stablyai, MIT-licensed. The full version-matched reference is served by the orca binary itself; this skill is the discovery stub that tells the agent how to load it. Repository: https://raw.githubusercontent.com/stablyai/orca/c9c6a00449eb72c41234940aead9832b8c46d030/skills/orchestration/SKILL.md Pinned version: install the code at commit c9c6a00449eb72c41234940aead9832b8c46d030 exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "orchestration". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. install the Orca runtime and resolve which executable to use — ORCA_DEV_REPO_ROOT, orca-ide, or orca — per the skill's instructions). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- c9c6a00 · 2026-10-13
- Sandbox install test
- Install ran in our test sandbox and failed · Nov 18, 2026 — Needs pnpm, npm cannot install it · Observed during the install test: registry.npmjs.org
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.