What is ai-dispatch CLAUDE.md?

A set of project instructions for developing, installing, releasing, and maintaining the ai-dispatch command-line tool.

Use it as a reference when installing ai-dispatch, signing its binary on macOS, preparing releases, running release dry-runs, and maintaining official guides.

It prevents contributors from using release steps that the project considers invalid or unsafe, such as manually creating tags or changing versions.

How agent-tools-org/ai-dispatch configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to agent-tools-org/ai-dispatch. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/agent-tools-org/ai-dispatch/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/agent-tools-org/ai-dispatch

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/agent-tools-org/ai-dispatch/claude-md/github.svg)](https://agentmods.dev/instructions/agent-tools-org/ai-dispatch/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/agent-tools-org/ai-dispatch/claude-md"><img src="https://agentmods.dev/badge/instructions/agent-tools-org/ai-dispatch/claude-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/agent-tools-org/ai-dispatch/claude-md"><img src="https://agentmods.dev/badge/instructions/agent-tools-org/ai-dispatch/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 4,800 This file is loaded in full into every session.

When invoked 4,800 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.04800	$0.04800
Opus 5.5	$0.01920	$0.01920
Sonnet 5.5	$0.00960	$0.00960
Haiku 4.5	$0.00480	$0.00480

Measured 17d ago against content hash f8244353e0b7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ai-dispatch CLAUDE.md 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 17d 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.md · 404 lines

How it starts

The opening of the file, as written. The whole thing — 404 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ai-dispatch (aid)

Install

NEVER cp binary to /opt/homebrew/bin/ — macOS provenance xattr blocks execution
/opt/homebrew/bin/aid is a symlink to ~/.cargo/bin/aid

Install command (MUST re-sign after copy — sandbox provenance blocks execution):

cp "$CARGO_TARGET_DIR/release/aid" ~/.cargo/bin/aid && codesign --force --sign - ~/.cargo/bin/aid

Release

Release must go through scripts/release.sh. Do not manually bump Cargo.toml, edit the top release entry in CHANGELOG.md, create the release commit, create the release tag, or push the release branch/tag by hand.

Tests and their compilation run on the build box through rbox, never on this Mac (boss rule 2026-09-11). Set AID_RELEASE_TEST_CMD='scripts/remote-test.sh' for releases; project verify is scripts/remote-test.sh -- --bin aid. AID starts verify outside the agent sandbox, inheriting the operator's AID_BUILD_BOX, RBOX_CONFIG, and PATH with rbox.

export AID_BUILD_BOX='<configured rbox name>'   # never commit the name
export AID_RELEASE_TEST_CMD='scripts/remote-test.sh'

The wrapper uses one rbox exec and runs the full workspace suite as the configured non-root user, with no test skips. Checkouts live at ~/.rbox/work/<repo-name>/<checkout-id>: the Git common directory supplies the repo name, and the branch (short HEAD SHA when detached) is sanitized to letters, digits, _, and -. All worktrees share $HOME/.rbox/target/<repo-name>. Rbox owns sync, transport, streaming, and the box lock. Use --jobs N (default 4, or AID_BUILD_JOBS), --timeout S (5400), --lock-timeout S (3600), -- <extra cargo test args>, or --dry-run to print the command. Exit 124 means the wait timed out and the job continues; 75 means the box lock was not acquired; 66 means no tests matched the filter. Diagnostics name the job ID when assigned (sync lock failure has no job yet). Completed test exits retain their status and are identified by rbox's completion marker. If the agent sandbox cannot reach Tailscale, stop at dry-run/fake-rbox checks; the operator runs the live release-test path. Never fall back to local compilation.

Read the full file on GitHub · 404 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.

17d ago Changed · -27 tokens per session f8244353e0b7
29d ago Changed · +1 lines · +10 tokens per session 622b9cd62716
1mo ago Changed · +3 lines · +34 tokens per session 440a64904b70
1mo ago Changed · +39 tokens per session 5d94f2333978
1mo ago Changed · +1 lines · +35 tokens per session 695bb2f65e2f
1mo ago Changed · +23 lines · +409 tokens per session 982d12f02a27
2mo ago Changed · +1 lines · +74 tokens per session 5894fe37c959
2mo ago First seen · 375 lines · 4,226 tokens per session scan A 24a16d1a9d6f

Subscribe to this mod's changes

ai-dispatch CLAUDE.md is an instructions file published in the GitHub repository agent-tools-org/ai-dispatch (12 stars, last pushed 2d ago), licensed MIT. It adds 4,800 tokens to every session, about $0.0192 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 2026-08-30.

ai-dispatch CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ai-dispatch CLAUDE.mdagent-tools-org/ai-dispatch	Instructions file	12	2d ago	MIT	A	4,800	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

ai-dispatch: Instructions file for Claude Code