cc-fleet Parallel Subagent Dispatch
Delegates one-shot or flat parallel batches of work to cc-fleet provider models (DeepSeek/GLM/Kimi/Qwen/MiniMax/Codex/Claude) via the headless `cc-fleet subagent` CLI.
AutomationAdvanced★ 213⑂ 29AI score 8/10Last updated: Aug 19, 2026
What it does
- Runs
cc-fleet subagentto execute provider models headless and return a JSON envelope straight on stdout — either a single shot or a lock-free parallel fan-out of independent tasks. - Encodes the provider selection ladder (user's pick → configured default → ask once), model tiers (
fastvsstrong), and guardrails:--timeout,--max-budget-usd,--max-turns,--prompt-file,--label. - Ships a full
error_codedispatch table so balance exhaustion, invalid keys, rate limits, timeouts, max-turns and oversized output each get the correct response instead of blind retries. - Covers long-running work via
--backgroundplus chunkedsubagent-status --waitpolling, and short follow-ups via--resume <session_id>.
Who it's for
- Users who already have cc-fleet installed with provider API keys configured.
- Engineers who want per-file or per-topic work spread across cheap models in parallel.
- Codex/Claude Code operators offloading heavy research so the lead session stays light.
Examples
- Summarize
docs/a.md,docs/b.md,docs/c.mdconcurrently on GLM and DeepSeek, then aggregate the results. - Delegate worst-case complexity analysis of
src/sort.gotodeepseek --model strongand consume only.result. - Run a single synthesis node with
cc-fleet subagent claude --model opusover gathered notes (mindful that it burns your own subscription window).
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/ethanhq/cc-fleet/HEAD/codex-plugin/skills/subagent/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the codex-plugin/skills/subagent folder from the GitHub repo ethanhq/cc-fleet into my ~/.claude/skills/subagent/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/ethanhq/cc-fleet.git /tmp/cc-fleet && mkdir -p ~/.claude/skills && cp -r /tmp/cc-fleet/codex-plugin/skills/subagent ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Confirm the cc-fleet CLI is installed and
cc-fleet doctor --jsonis clean (a working Claude Code binary is required). - Clone the repo:
git clone https://github.com/ethanhq/cc-fleet.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r cc-fleet/codex-plugin/skills/subagent ~/.claude/skills/ - Also copy the shared docs the skill references:
cp -r cc-fleet/codex-plugin/skills/cc-fleet-shared ~/.claude/skills/ - Restart Claude Code and try: "use a cc-fleet subagent fan-out to summarize these files in parallel".
- If no provider exists yet, add one with
cc-fleet add <provider>and pin it withcc-fleet default <provider>.
View source on GitHub ↗License: Apache-2.0