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.
npx agentmods add skills/a1f/agent-templates/dispatchnpx skills add a1f/agent-templates --skill dispatchgit clone --depth 1 https://github.com/a1f/agent-templatesWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00045 | $0.02188 |
| Opus 5 | $0.00023 | $0.01094 |
| Sonnet 5 | $0.00009 | $0.00438 |
| Haiku 4.5 | $0.00005 | $0.00219 |
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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dispatch
Take the work that is ready to start in a plan and fan it out: one tmux window per
PR, each in its own worktree, each running a coding agent on /make-pr <ref> <PR#>.
/dispatch [<issue ref | artifact | design doc>] [--pr <PR#> ...] [--session NAME] [--dry-run]
1. Find the unblocked PRs
2. Confirm which to schedule (gate — always)
3. Write the plan
4. Fan out — one window per PR
You own steps 1–3 (judgement: what is ready, what the user confirms, the plan). The script owns step 4 — the mechanics (worktree, branch, window, launch) — so they are deterministic and repeatable.
--pr <PR#> schedules just those PRs (repeat the flag for a few, e.g.
--pr 1.5 --pr 1.7) instead of the whole ready set — use it when you want one PR, not the
batch. It preselects; it does not skip the confirmation gate. You still resolve each
named PR from the source (for its What → slug and its ref) and still show the rest of
the ready list so the user can add to the selection.
Phase 1 — Find the unblocked PRs
Read the source the user named:
- Issue (
#N,123, or a URL) —gh issue view <N> --json title,body -q .body. - File — read it. Artifact URL — fetch it.
- Nothing given? Scan the conversation for the last
PRD published: ... (issue #N)or a plan the user just approved. Confirm the number with the user before proceeding.
Find the PR rows (a ## PR breakdown section from /pr-breakdown has them as
| PR | What | LOC | Done when |). If the source has no PR-level breakdown, stop
and point the user at /pr-breakdown — do not invent one.
Then decide which rows are ready to start:
| Signal | Where it comes from |
|---|---|
| Dependencies satisfied | the plan's own ordering and parallelism notes (*n.1 ∥ n.2 (independent)* = no dependency between them; otherwise later rows in a slice wait on earlier ones) |
| Not already done | gh pr list --state all --search "<PR#>" — skip rows with a merged or open PR |
| Not already started | git branch -a and git worktree list — skip rows that already have a branch or worktree |
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.
- yesterday First seen · 169 lines · 45 tokens per session scan A 3e8a832ecc40
dispatch is a skill published in the GitHub repository a1f/agent-templates (2 stars, last pushed yesterday), licensed MIT. It adds 45 tokens to every session and 2,188 once invoked, about $0.0002 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…