dispatch

A command that coordinates several coding sub-agents at the same time using separate work slots. Some slots handle new development, while others respond to failures or stalled reviews.

In plain words
What is it for?
Use it to start an orchestration session with a chosen number of total slots or with explicit development, rescue, and flexible slot counts.
Why use it?
It keeps development work moving while reserving capacity for urgent problems such as broken continuous integration or unresolved pull-request discussions.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/skyfox675/agents-skills/dispatch
Clone the repo
git clone --depth 1 https://github.com/skyfox675/agents-skills

Made for: Claude Code.

Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,182 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00069 $0.01182
Opus 5 $0.00034 $0.00591
Sonnet 5 $0.00014 $0.00236
Haiku 4.5 $0.00007 $0.00118

Measured 2d ago against content hash 72d845ef413b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dispatch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/dispatch.md · 39 lines

How it starts

The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/dispatch — fill orchestration slots

Operate in caveman mode (load the caveman skill) — orchestration is high-volume and tokens compound across rounds. Keep gh commands, labels, lock markers, and slot/PR numbers byte-exact; compress only the narration.

Arguments: $ARGUMENTS

Parse them as:

  • A single number N is the total slot count. Default split: rescue = max(1, ⌊N/8⌋), flex = ⌊N/4⌋, dev = the remainder. /dispatch 8 → dev:5 rescue:1 flex:2.
  • dev:<n>, rescue:<n>, flex:<n> tokens (anywhere in the args) set counts explicitly and always win over the derived split. Any subset is fine — unallocated remainder from a total goes to dev. /dispatch 8 rescue:2 → dev:4 rescue:2 flex:2.
  • No arguments is an error: ask the operator for a slot count rather than guessing — slot capacity is operator policy (see orchestrating-slots).

Slot semantics (full protocol in the orchestrating-slots skill — read it before acting):

  • A dev slot runs a sub-agent implementing a groomed issue from the queue. Dedicated: it refills with dev work when freed.
  • A rescue slot is held for inbound incidents: CI failures on owned PRs, bot-review thread sweeps, DIRTY/stuck PRs, deployment or smoke-check failures. Dedicated: never backfill rescue capacity with dev work just because it's idle — rescue capacity exists so a queue incident never waits behind feature work.
  • A flex slot works in both directions: fill it with dev work when the pipeline is healthy, but it is the first capacity reclaimed when incidents exceed rescue capacity — preempt its dev dispatch only if the work isn't yet near a push (otherwise let it finish, then convert). Flex slots are also the first to go idle under backpressure.
  • A slot is held from claim-lock placement until the linked PR closes. Existing in-flight work you own counts against today's slots — reconcile before dispatching anything new.

Steps

  1. Resolve identity and reconcile. ME=$(gh api user --jq .login). Survey open PRs by ownership: yours (each holds a dev or flex slot already), other operators' (their files become the conflict-zone list passed into every brief — never dispatch onto them), unassigned. Release any stale claims per the issue-locking skill (gh-issue-locking / jira-issue-locking).

Read the full file on GitHub · 39 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 39 lines · 69 tokens per session scan A 72d845ef413b

Subscribe to this mod's changes

dispatch is a command published in the GitHub repository skyfox675/agents-skills (10 stars, last pushed 1mo ago), licensed MIT. It adds 69 tokens to every session and 1,182 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.