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/nikzlabs/shipit/claude-mdgit clone --depth 1 https://github.com/nikzlabs/shipitWrote 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/nikzlabs/shipit/claude-md)<a href="https://agentmods.dev/instructions/nikzlabs/shipit/claude-md"><img src="https://agentmods.dev/badge/instructions/nikzlabs/shipit/claude-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 | $0.10578 | $0.10578 |
| Opus 5 | $0.05289 | $0.05289 |
| Sonnet 5 | $0.02116 | $0.02116 |
| Haiku 4.5 | $0.01058 | $0.01058 |
Grade A, and why
shipit 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 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 — 297 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
ShipIt is a browser-based AI editor — describe what you want in chat, the agent writes the code, and you see results live. The agent runs as a CLI inside a session container; Claude Code CLI is the default backend, Codex CLI is also supported, and the architecture is agent-agnostic so additional backends can be added later. Authentication normally uses the user's existing subscription with the chosen provider; a metered API key exists only as a fallback ranked below connected accounts.
Product principles
These govern what ShipIt is. They override convenience, "what other tools do," and "this is how the underlying platform works." A proposal that conflicts with one is wrong. Design docs cite them by number — §1–§5 are stable identifiers; compress the prose, never renumber.
1. ShipIt is the surface. The user does not leave it.
You build, review, ship, and debug inside one chat-shaped IDE. PRs, CI status, deploy status, diffs, commit history, conversation history, terminal output, preview, and merge conflicts all surface inline — no GitHub tab, hosting dashboard, CI tab, or local terminal required. Sending the user elsewhere is a failure of the product, not a feature.
2. Inline beats link-out. Always.
If the upstream system has the data, ShipIt fetches and renders it. Links to GitHub or the cloud provider are escape hatches in overflow menus, never the happy path — "View on GitHub" sits beside the PR card, and we never bounce anyone to GitHub's diff viewer. The reason is compounding: once the user is reading a PR on github.com, the next comment, re-request, and fixup happen there too. The cycle has to start somewhere; we keep it inside.
3. External tabs are reserved for things ShipIt does not own.
The whole list: OAuth / auth flows (Anthropic, GitHub own their login screens), account and billing pages (provider billing, GitHub repo creation and settings), and external documentation the user explicitly clicks through to. "The PR was created so let's open it" is not on it.
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 Changed · +177 tokens per session 202180354c5d
- 4d ago First seen · 297 lines · 10,401 tokens per session scan A a8d1aee6dd1e
shipit CLAUDE.md is an instructions file published in the GitHub repository nikzlabs/shipit (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 10,578 tokens to every session, about $0.0529 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
KDD AGENTS.md
AGENTS.md instructions for MauricioPerera/KDD: Este repositorio usa la metodología KDD (Knowledge-Driven Development). Las reglas obligatorias completas están en .agents/AGENTS.md y la skill local en .agents/skills/kdd-okf-ccdd-hybrid/SKILL.md.
AmbyKit CLAUDE.md
Instructions for ambystechcom/AmbyKit, covering claude code — notes for the ambykit repo and claude-specific.
lumen AGENTS.md
AGENTS.md instructions for ory/lumen, covering agents.md, repository expectations and commands.
ai-auto-work CLAUDE.md
Claude Code instructions for chaohong-ai/ai-auto-work, covering gamemaker - ai native 游戏创作平台, 项目简介, ai 工作目录约定 and 按需加载（不默认装载）.
catalyst AGENTS.md
AGENTS.md instructions for coalesce-labs/catalyst, covering agents.md, what this repository is, agent philosophy, build & test and key principles.
truthmark CLAUDE.md
Claude Code instructions for merlinhu1/truthmark, covering authority, documentation scope, product boundary, rules and instruction surface boundary.