dispatch-grok
Skill24 starsChanged 45 days agoMITView on GitHub
Dispatch one or more tasks to grok (Grok Build) agents in herdr workspaces — the orchestrator designs each lane's plan and acceptance criteria, grok pursues them — supervise them to completion, then push each lane and open its pull request. Use when the user asks to dispatch/fan out/派发 tasks to grok agents or herdr lanes, to run several grok tasks in parallel worktrees, or to resume supervision of an existing dispatch-grok run (`--resume`).
- Commits and pushes
3 other projects use a version of this.
Adopted by
Most recently taken up by
- KUN1007/kun-galgame-patch-nexttheir own version, first seen Oct 8, 2026
- KUN1007/kun-galgame-nuxt3their own version, first seen Oct 8, 2026
- KUN1007/kun-galgame-nuxt4their own version, first seen Oct 8, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- KUN1007/kun-galgame-nuxt3★ 2.3ksame name
- KUN1007/kun-galgame-nuxt4★ 2.3ksame name
- KUN1007/kun-galgame-patch-next★ 345same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository3 other files it gives its agents.
- dispatch-antigravitySkill
- dispatch-codexSkill
- dispatch-opencodeSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 0 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 44
Judge a lane from disk, not from the screen: Every grok session keeps its own directory under ~/.grok/sessions/, and two small files in it …
- Very specific: line 71
A lane's
chat_history.jsonlgrows without bound: Never parse one whole — prefer signals.json and summary.json, and read only the tail of … - Very specific: line 78
Run test "${HERDR_ENV:-}" = 1, test -n "${HERDR_PANE_ID:-}" and herdr agent list.
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Dispatch tasks to grok agents running under herdr
The raw request is the text passed to this skill — its arguments, or, when it was invoked with none, the task text in the user's own message.
You are an orchestrator. You do not implement the tasks yourself. You split the request into
lanes, give each lane its own herdr workspace and its own grok agent, and then keep those agents
alive and moving until every lane genuinely finishes. You do the planning; grok does the work:
each lane gets a plan and acceptance criteria you author and the user confirms (§3), and — when
this machine has grok's goal mode enabled — runs as a grok goal (/goal, §5c), an objective
grok works across rounds and only marks complete after its own evidence review. A lane is not
finished when its agent stops — it is finished when its work is verified, its branch is pushed, and
its pull request is open (§6f).
How to read this skill
The procedure is split across four files, and the section numbers (§0–§8) are continuous across all of them, so a cross-reference means the same thing wherever you are:
| Sections | File | Read it |
| §0 invariants, §1 gate and parse | this file | always, and §0 again at the start of every sweep |
| §2–§4, §5b | references/plan.md | on a fresh dispatch, before creating anything |
| §5a, §5c, §6a, §6c, §6d, §6g, §6h | references/driver.md | everything grok-specific: launch, probe, classify, steer |
| §6b, §6e, §6f, §6i, §6j, §7, §8 | references/supervise.md | before the first supervision sweep |
references/plan.md and references/supervise.md are symlinks into the plugin's skills/_shared/:
the shared halves stay single-sourced while still resolving when this skill's directory is copied on
its own, which is how the skills.sh installer places it. Both are read by path,
not invoked.
§0 Invariants — reread every sweep, never work from memory
- The state file is the only truth.
~/.claude/dispatch-grok/<run-id>/state.json. Begin every sweep by reading it; your conversation memory may have been compacted away. Rewrite it atomically (write.tmp, thenmv) at the end of every sweep. - Judge a lane from disk, not from the screen. Every grok session keeps its own directory under
~/.grok/sessions/, and two small files in it —signals.json(context-window usage as a percentage, compaction count, turn count, error and tool-failure counts) andsummary.json(model, message count, last activity, git branch) — answer most of what a sweep needs without parsing a transcript at all.events.jsonlsupplies turn boundaries and permission requests. Terminal output is the fallback, never the primary signal. agent_statusalone never means "finished", and it can lie outright. A background lane reportsdone(notidle) when unseen work ends — but it reads the same when the prompt was swallowed, when grok froze, and when herdr misclassified a modal. Always corroborate (§6). The same goes for a lane's notify-back or help ring (§5b): it is a doorbell that starts a sweep sooner, never evidence that skips §6e.- Grok's goal state is not on disk. Unlike its context and turn counters, a goal's
active/paused/blockedstatus lives only inside the running session and is read by asking/goal statusin the pane — which costs a turn. So the sweep never polls it routinely: it asks only when it is already about to act on that lane (§6cstalled/idle_incomplete), and until it asks, goal state isunknownand proves nothing either way. - Never touch what you did not create. Act only on ids recorded in the state file. Never
herdr server stop. Neverherdr agent focus/workspace focus— it steals the human's UI focus and silently flipsdonetoidle, destroying your own signal. Nevergrok leader kill: one leader process backs every grok session on this machine, including the user's own. - Never destroy work; publish only what you verified. Finishing a lane means pushing its own
branch and opening a PR for it (§6f) — both additive and reversible, and both yours to do, never
the lane's. Everything else stays forbidden: never force-push (
--force,--force-with-lease), never push the base branch or any branch absent from the state file, never merge, neverworktree remove, nevergrok sessions delete. Print those commands and let the user run them:worktree removekills the running grok process and deletes uncommitted changes even without--force. - A lane's
chat_history.jsonlgrows without bound. Never parse one whole — prefersignals.jsonandsummary.json, and read only the tail ofevents.jsonl.
§1 Gate and parse
Run test "${HERDR_ENV:-}" = 1, test -n "${HERDR_PANE_ID:-}" and herdr agent list. If
HERDR_ENV or HERDR_PANE_ID is unset or the CLI cannot reach the socket, stop and tell the user in
Chinese that this session is not inside a herdr pane, so there is nothing to dispatch into — an
exported HERDR_ENV alone can pass in a non-pane shell, and a run recorded without its pane id has
no working notify-back. Do not install or launch herdr, and do not run grok yourself.
Parse flags from the raw request; everything else is task text.
| Flag | Meaning | Default |
--lanes N | cap on concurrent lanes, 1–16 | 16 |
--base <ref> | base ref for lane branches | origin/<current> if it exists, else current branch |
--no-yolo | launch each lane with --permission-mode default so tool calls prompt for approval | off — yolo is the default |
--yolo | accepted and explicit, but redundant: this is already the default | on |
--draft | open pull requests as drafts instead of ready for review | off — ready for review is the default |
--no-pr | push each verified lane but stop there; print the gh pr create command instead | off |
--resume | skip §2–§5 (this gate and parse still run); run ONE supervision sweep over the existing state file | off |
--no-loop | do not arm the recurring supervision loop after dispatch | off |
--no-yolo must be passed through, not merely un-passed. ~/.grok/config.toml can turn yolo on
globally ([ui] yolo / [ui] permission_mode), so omitting the bypass flag does not restore
prompting — the lane would still auto-approve everything while the §3 summary claimed otherwise.
Under --no-yolo the driver passes --permission-mode default explicitly (§5c). Read the config
before summarising the posture in §3, and describe what will actually happen.
Every lane gets its own git worktree. This is not a flag and there is no opt-out. With approvals
bypassed by default (§5c), the worktree boundary is the only thing left keeping one lane's mistakes
out of the other lanes and out of the user's own checkout. If the request contains --no-worktree,
stop before creating anything: say in Chinese that this skill always isolates lanes in worktrees
and that the flag no longer exists, and ask the user to re-run without it. Grok's own --worktree /
-w flag is likewise never used — §4 explains why.
With --resume, first locate the run — conversation memory may be gone (§0.1): scan
~/.claude/dispatch-grok/*/state.json for runs whose repo matches the cwd and that still hold
non-terminal lanes; one match sweeps it, several means ask the user which, none means say so and
stop. Then read references/driver.md and references/supervise.md and go to §6. With no task text
and no --resume, ask the user in Chinese what to dispatch, and stop.
Otherwise — a fresh dispatch — read references/plan.md now and continue at §2.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

