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 skills add GregoryHo/arcforge --skill dispatchinggit clone --depth 1 https://github.com/GregoryHo/arcforgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gregoryho/arcforge/dispatching)<a href="https://agentmods.dev/skills/gregoryho/arcforge/dispatching"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/dispatching/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gregoryho/arcforge/dispatching"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/dispatching.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00057 | $0.01928 |
| Opus 5.5 | $0.00023 | $0.00771 |
| Sonnet 5.5 | $0.00011 | $0.00386 |
| Haiku 4.5 | $0.00006 | $0.00193 |
Grade A, and why
dispatching 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 today.
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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dispatching
Parallel work fails in four places, in order: the split was wrong, the workspaces overlapped, the brief was incomplete, or the report was believed. Each step below closes one of them.
Splitting work you could have done yourself in one pass costs more than it saves. Dispatch when the pieces genuinely do not need each other, or when one of them needs a fresh context to be judged honestly.
Step 1 — Prove the pieces are independent
Independent means all three: no shared dependency, no shared file, and each piece understandable without reading the others. Two failures in the same module, or a fix that might turn out to fix the others too, are one piece — dispatch them as one.
Write the split down before dispatching: which piece owns which files. A piece whose file list you cannot state is not scoped yet.
- Done when every piece has a file list and no file appears on two lists.
Do not pre-negotiate conflicts beyond that list. Predicting which lines will collide is guesswork that costs more than resolving the collision later.
Step 2 — Give every writer its own worktree
Two agents writing into one checkout will overwrite each other's edits and stage each other's files. Isolation is not a precaution here, it is the precondition: if the pieces write, each gets its own worktree.
arcforge worktree add <name> [--branch <b>] [--from <ref>] [--setup] --json
arcforge worktree list --json
arcforge worktree remove <name> [--force]
Read the path field out of the JSON. The location is derived by the engine, so
a path you construct, remember, or guess will be the wrong one and everything
downstream that looks for the worktree will miss it. --setup runs the project's
installer inside the new tree; list annotates each entry by kind, and cd to a
path is how you enter one — there is no switch command.
Already inside a worktree? Work there. Never nest a worktree inside a worktree. A path the user names explicitly overrides all of this — honor it with raw git.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today First seen · 163 lines · 57 tokens per session scan A 3e16bd233ceb
dispatching is a skill published in the GitHub repository GregoryHo/arcforge (8 stars, last pushed 12d ago), licensed MIT. It adds 57 tokens to every session and 1,928 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-07.
dispatching compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dispatchingGregoryHo/arcforge | Skill | 8 | 12d ago | MIT | A | 57 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |