agentmods.dev

flue CLAUDE.md

Repository instructions for working on the flue project. They define its branch-and-pull-request workflow, package manager, and required checks for web and Go code.

In plain words
What is it for?
Use them when changing the project, especially to choose pnpm instead of npm, work on a branch, run the relevant Vitest, lint, Go, or cross-cutting tests, and submit a pull request.
Why use it?
They help avoid incorrect dependency files and missed tests or lint checks in a project that contains a web workspace and Go code.

Instructions file

10 6d ago A scan 351 tokens original MIT
View the source file karnstack/flue
Install

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.

agentmods
npx agentmods add instructions/karnstack/flue/claude-md
Clone the repo
git clone --depth 1 https://github.com/karnstack/flue
Per session 351 This file is loaded in full into every session.
When invoked 351 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00351 $0.00351
Opus 5 $0.00176 $0.00176
Sonnet 5 $0.00070 $0.00070
Haiku 4.5 $0.00035 $0.00035

Measured 2d ago against content hash 5ff955a31706, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flue 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 2d 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 · 23 lines

What it actually says

flue — agent notes

Workflow

  • main is protected. Never commit to or merge into main locally.
  • All work happens on a branch (worktrees under .claude/worktrees/), and lands via a pull request — always, even for one-line fixes. Push the branch, open the PR with gh pr create, and hand the URL back.
  • pnpm only. web/ is a pnpm workspace (pnpm-lock.yaml, pnpm-workspace.yaml) and there is no package-lock.json. Use pnpm install, pnpm add, pnpm run <script>, and pnpm dlx for one-off binaries. Never npm, npx or yarn: npm install writes a package-lock.json and a flat node_modules beside the pnpm store, and npm's allow-scripts gate leaves esbuild's postinstall unrun, so the vite build in styles.build.test.ts fails for reasons that look nothing like the cause.
  • Run the affected test surface before pushing: cd web && pnpm vitest run and pnpm run lint for web changes, go test ./... for Go, make test for anything cross-cutting. web's styles.build.test.ts is a real vite build and is the guard for the Tailwind prose scanner — comments in scanned web sources can compile stray CSS rules; see the notes at the top of web/src/styles.css before writing prose there.
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.

  1. 2d ago First seen · 23 lines · 351 tokens per session scan A 5ff955a31706

Subscribe to this mod's changes

flue CLAUDE.md is an instructions file published in the GitHub repository karnstack/flue (10 stars, last pushed 6d ago), licensed MIT. It adds 351 tokens to every session, about $0.0018 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.

Code reviewGit and workflowTestingaiclaude-codecodexmultiple-devicesremote-controlterminal
Related

Other instructions, from other repositories

terminal-browser AGENTS.md

Instructions for zenbu-labs/terminal-browser, a project described as: A browser inside your terminal.

zenbu-labs/terminal-browser · 304 tokens

obsidian-claude-sidebar CLAUDE.md

Instructions for derek-larson14/obsidian-claude-sidebar, covering claude sidebar - agent guide, what this plugin does, two things you can help the user do, step 1: diagnose and step 2a: file a bug report.

derek-larson14/obsidian-claude-sidebar · 2,409 tokens

liney AGENTS.md

Instructions for everettjf/liney, covering liney repository collaboration guide, 交流语言 / communication language, project overview, repository layout and build and test.

everettjf/liney · 1,589 tokens

waveloom AGENTS.md

Instructions for Menfre01/waveloom, covering waveloom, 项目概要, 编码规范, 代码审查 and 开发流程.

Menfre01/waveloom · 3,069 tokens

superset AGENTS.md

AGENTS.md instructions for superset-sh/superset, covering superset monorepo, project structure, exception: shadcn/ui components, database and releases.

superset-sh/superset · 2,316 tokens

starboard CLAUDE.md

Instructions for palamim/starboard, covering claude.md, commands, code style, architecture and non-obvious gotchas (not discoverable by reading the code).

palamim/starboard · 1,892 tokens