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 instructions/yunyoujun/craft-hub/agents-mdgit clone --depth 1 https://github.com/YunYouJun/craft-hubWrote 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/instructions/yunyoujun/craft-hub/agents-md)<a href="https://agentmods.dev/instructions/yunyoujun/craft-hub/agents-md"><img src="https://agentmods.dev/badge/instructions/yunyoujun/craft-hub/agents-md.svg" alt="Measured on agentmods" 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 | $0.00492 | $0.00492 |
| Opus 5 | $0.00246 | $0.00246 |
| Sonnet 5 | $0.00098 | $0.00098 |
| Haiku 4.5 | $0.00049 | $0.00049 |
Grade A, and why
craft-hub AGENTS.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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Craft Hub agent guide
Craft Hub is a local, cross-project developer workbench. Project Palette discovers project commands and agent skills; the shared runtime powers the CLI, browser UI, and Electron shell.
Architecture
packages/craft-hub: source of truth for project registry, capability discovery, trust, execution, persistence, local API, and CLI.apps/web: Vue UI that consumes the local API. Keep it usable in a normal browser.apps/desktop: thin Electron host. Keep domain logic out of this package.examples/sample-project: deterministic manual-test fixture.docs/design/workbench-concept.webp: visual baseline for the workbench.
Keep the runtime vendor-neutral. Agent integrations belong behind adapters; editor or desktop clients consume public runtime contracts.
Keep runtime code, fixtures, examples, documentation, and tests free of organization-specific product, account-system, package-scope, registry, editor, and intranet-domain names. Those belong only in downstream distributions. Run pnpm check:public-boundary; CI treats any denylist match as a hard failure.
Safety boundary
Capability discovery is read-only. A new project is untrusted, and command execution requires explicit trust. Represent execution as command plus args, set shell: false, and preserve the project working directory. Add raw shell support only through an explicit future config surface with its own review.
Persist user state in the operating-system Craft Hub data directory. The project repository only owns optional .craft-hub/project.jsonc metadata. Keep it declarative, validate it through the Zod schema, preserve comments with JSONC AST edits, and place third-party data under extensions.
Working conventions
- Use pnpm catalog dependencies from
pnpm-workspace.yaml. - Use ESM and strict TypeScript. Add JSDoc to public runtime APIs.
- Follow the repository ESLint config: no semicolons and single quotes.
- Add discovery and trust/execution tests with every runtime behavior change.
- Run
pnpm lint,pnpm typecheck,pnpm test --run, andpnpm buildbefore handoff.
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.
- 4d ago First seen · 32 lines · 492 tokens per session scan A 5771a9b9cb23
craft-hub AGENTS.md is an instructions file published in the GitHub repository YunYouJun/craft-hub (1 stars, last pushed 3d ago), licensed MIT. It adds 492 tokens to every session, about $0.0025 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 instructions, from other repositories
stashbase AGENTS.md
AGENTS.md instructions for liliu-z/stashbase, covering agent maintenance contract for this repo, github access for this repository, temporary worktrees, electron launch environment and documentation route.
chat-on-steroids CLAUDE.md
Claude Code instructions for totec448-spec/chat-on-steroids: Read and follow AGENTS.md before changing this repository.
tandem-browser CLAUDE.md
Claude Code instructions for hydro13/tandem-browser: Read AGENTS.md in full — it is the development guide for coding agents working on this repository: workflow, testing gates, git discipline, context discipline, and the anti-detection architecture rules that must never be violated.
Knote AGENTS.md
Instructions for 1661169091kiwi/Knote, covering agents.md, 构建与验证约定（用户明确要求，所有会话必须遵守） and 测试说明.
clodex-ide AGENTS.md
AGENTS.md instructions for mereyabdenbekuly-ctrl/clodex-ide, covering clodex repository agent rules, open/closed product boundary, commercial product invariant, current protocol v0 phase and current authorization state.
braid AGENTS.md
Instructions for openbraid/braid, covering agents.md, what braid is, repository layout, the two modes — read this before touching storage and migration code that looks like a botched rename.