agentcn: Skill for Claude Code

.agents/skills/flue/SKILL.md

flue is a skill for Claude Code, Codex from shadcn-labs/agentcn. It costs 55 tokens per session (1,786 once invoked), scanned A, original, MIT.

A set of instructions for building and maintaining Flue agents and workflows. Flue is a tool for defining coding agents, reusable actions, communication channels, storage, routing, and monitoring.

In plain words
What is it for?
Use it to build, debug, review, or document Flue agents, workflows, channels, skills, tools, storage, routing, and command-line usage.
Why use it?
It directs coding agents to documentation for the installed Flue version, reducing the risk of following outdated or incorrect instructions.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; installed under .agents/ (shared by several agents).

This is shadcn-labs/agentcn's own configuration. It tells Claude Code and Codex how to work on agentcn itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agentcn configures →

Reuse

Borrowing it

Nothing to install: this file belongs to shadcn-labs/agentcn. 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/shadcn-labs/agentcn/main/.agents/skills/flue/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/shadcn-labs/agentcn

Made for: Claude Code, Codex.

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.

agentmods badge for flue

README.md
[![agentmods](https://agentmods.dev/badge/skills/shadcn-labs/agentcn/flue/github.svg)](https://agentmods.dev/skills/shadcn-labs/agentcn/flue)
Your own site
<a href="https://agentmods.dev/skills/shadcn-labs/agentcn/flue"><img src="https://agentmods.dev/badge/skills/shadcn-labs/agentcn/flue/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.

agentmods 80×15 button for flue

Your own site · 80×15
<a href="https://agentmods.dev/skills/shadcn-labs/agentcn/flue"><img src="https://agentmods.dev/badge/skills/shadcn-labs/agentcn/flue.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,786 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Prompt Injection · line 14
    Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
    Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
How audits are shown
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 5.1 $0.00055 $0.01786
Opus 5 $0.00028 $0.00893
Sonnet 5 $0.00011 $0.00357
Haiku 4.5 $0.00006 $0.00179

Measured 9d ago against content hash e218dec14742, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

flue 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 9d 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.

.agents/skills/flue/SKILL.md · 165 lines

How it starts

The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Flue

Use flue docs to read the documentation bundled with the installed @flue/cli version. Choose relevant paths from the catalog below and run flue docs read <path>. If no catalog entry matches your task, run flue docs search <query>, then read the most relevant result with flue docs read <path>.

For example, flue docs search "durable execution" searches with the query durable execution. If it returns the path concepts/durable-execution, run flue docs read concepts/durable-execution to read that page.

Documentation Catalog

api/action-api -- Action API
  Reference for defining reusable finite Actions with @flue/runtime.
api/agent-api -- Agent API
  Reference for defining agents and running agent operations with @flue/runtime.
api/data-persistence-api -- Data Persistence API
  Reference for Flue persistence adapters, stores, and session data.
api/errors-reference -- Errors Reference
  Reference Flue transport errors, runtime failures, and development diagnostics.
api/events-reference -- Events Reference
  Reference runtime activity, attached-agent event types, and global observation APIs.
api/provider-api -- Provider API
  Register custom model providers and override built-in provider transport.
api/routing-api -- Routing API
  Compose Flue routes in an authored application entrypoint.
api/sandbox-api -- Sandbox Adapter API
  Adapt a provider sandbox SDK into Flue's public sandbox contract.
api/streaming-protocol -- Streaming Protocol
  Reference for reading Flue agent and workflow event streams over Durable Streams.
api/workflow-api -- Workflow API
  Reference for creating and invoking workflows with @flue/runtime.
cli/add -- flue add
  Reference for discovering and applying Flue implementation blueprints.
cli/build -- flue build
  Reference for creating deployable Flue application artifacts.
cli/connect -- flue connect
  Reference for opening an interactive local agent-instance session from the command line.
cli/dev -- flue dev
  Reference for starting a watch-mode local Flue development server.
cli/docs -- flue docs
  Reference for listing, reading, and searching the bundled Flue documentation.
cli/init -- flue init
  Reference for creating an initial Flue project configuration file.
cli/overview -- CLI
  Use the Flue CLI to configure, develop, exercise, inspect, and build an application.
cli/run -- flue run
  Reference for executing one workflow invocation from the command line.
cli/update -- flue update
  Reference for updating integrations from newer Flue blueprint upgrade guides.
concepts/agents -- What is an agent?
  What an AI agent actually is, why a model alone isn't one, and what makes a Flue agent different.
concepts/durable-execution -- Durable Agents
  Understand how Flue agents and workflows handle server restarts, interrupted connections, and other disruptions.
ecosystem/channels/discord -- Discord
ecosystem/channels/github -- GitHub
ecosystem/channels/google-chat -- Google Chat
ecosystem/channels/intercom -- Intercom
ecosystem/channels/linear -- Linear
ecosystem/channels/messenger -- Facebook Messenger
ecosystem/channels/notion -- Notion
ecosystem/channels/resend -- Resend
ecosystem/channels/salesforce-marketing-cloud -- Salesforce Marketing Cloud
ecosystem/channels/shopify -- Shopify
ecosystem/channels/slack -- Slack
ecosystem/channels/stripe -- Stripe
ecosystem/channels/teams -- Microsoft Teams
ecosystem/channels/telegram -- Telegram
ecosystem/channels/twilio -- Twilio
ecosystem/channels/whatsapp -- WhatsApp
ecosystem/channels/zendesk -- Zendesk
ecosystem/databases/libsql -- libSQL
ecosystem/databases/mongodb -- MongoDB
ecosystem/databases/mysql -- MySQL
ecosystem/databases/postgres -- Postgres
ecosystem/databases/redis -- Redis
ecosystem/databases/supabase -- Supabase
ecosystem/databases/turso -- Turso
ecosystem/databases/valkey -- Valkey
ecosystem/deploy/aws -- Deploy Agents on AWS
ecosystem/deploy/cloudflare -- Deploy to Cloudflare
ecosystem/deploy/docker -- Deploy Agents with Docker
ecosystem/deploy/fly -- Deploy Agents on Fly.io
ecosystem/deploy/github-actions -- Build Agents for GitHub Actions
ecosystem/deploy/gitlab-ci -- Build Agents for GitLab CI/CD
ecosystem/deploy/node -- Deploy Agents on Node.js
ecosystem/deploy/railway -- Deploy Agents on Railway
ecosystem/deploy/render -- Deploy Agents on Render
ecosystem/deploy/sst -- Deploy Agents on SST
ecosystem/sandboxes/boxd -- boxd
ecosystem/sandboxes/cloudflare -- Cloudflare Sandbox
ecosystem/sandboxes/cloudflare-shell -- Cloudflare Shell
ecosystem/sandboxes/daytona -- Daytona
ecosystem/sandboxes/e2b -- E2B
ecosystem/sandboxes/exedev -- exe.dev
ecosystem/sandboxes/islo -- islo
ecosystem/sandboxes/mirage -- Mirage
ecosystem/sandboxes/modal -- Modal
ecosystem/sandboxes/vercel -- Vercel Sandbox
ecosystem/tooling/braintrust -- Braintrust
ecosystem/tooling/opentelemetry -- OpenTelemetry
ecosystem/tooling/sentry -- Sentry
ecosystem/tooling/vitest-evals -- Vitest Evals
getting-started/quickstart -- Getting Started
  Set up a Flue project automatically or create your first agent manually.
guide/building-agents -- Agents
  Create an agent, configure its capabilities, and send it messages over time.
guide/channels -- Channels
  Receive verified provider events and connect them to Flue applications.
guide/database -- Database
  Configure database-backed state for Flue agents and workflow runs.
guide/evals -- Evals
  Evaluate Flue agents with repeatable Vitest suites using vitest-evals.
guide/models -- LLM (Models & Providers)
  Select models, configure providers, and tune reasoning behavior in Flue agents.
guide/observability -- Observability
  Inspect workflow runs, monitor agent activity, and export telemetry from your application.
guide/project-layout -- Project Layout
  Understand the source files and generated output in a Flue project.
guide/react -- React
  Build React interfaces for live agent conversations and workflow runs.
guide/routing -- Routing
  Compose Flue with application routes, middleware, and custom HTTP ingress.
guide/sandboxes -- Sandboxes
  Give agents a workspace for files and command-driven work.
guide/schedules -- Schedules
  Invoke Flue workflows or dispatch agent input on a schedule with Cloudflare or Node.js.
guide/skills -- Skills
  Add Agent Skills to Flue agents and invoke them from sessions.
guide/subagents -- Subagents
  Let agents delegate focused work to named specialists.
guide/targets/cloudflare -- Cloudflare
  Understand the Cloudflare-specific runtime behavior and APIs for Flue applications.
guide/targets/node -- Node.js
  Understand the Node.js-specific runtime behavior and APIs for Flue applications.
guide/tools -- Tools
  Give agents application capabilities through custom tools and MCP servers.
guide/workflows -- Workflows
  Create finite agent-backed operations from inline or reusable Actions.
introduction/why-flue -- Why Flue?
  Build autonomous AI agents and powerful workflows with a programmable TypeScript harness, and run them anywhere.
reference/configuration -- Configuration
  Reference for flue.config.ts options.
sdk/agents -- client.agents
  Invoke persistent agent instances and stream their events.
sdk/client -- createFlueClient(...)
  Configure an SDK client for a deployed Flue application.
sdk/errors -- Errors
  SDK HTTP and stream error types.
sdk/events -- Events and records
  SDK event, workflow-run record, and normalized model-turn types.
sdk/overview -- SDK overview
  Reference for consuming deployed Flue agents and workflows with @flue/sdk.
sdk/runs -- client.runs
  Inspect and stream HTTP-exposed workflow runs.
sdk/workflows -- client.workflows
  Start workflow runs and receive their run ID.

Read the full file on GitHub · 165 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.

  1. 9d ago First seen · 165 lines · 55 tokens per session scan A e218dec14742

Subscribe to this mod's changes

flue is a skill published in the GitHub repository shadcn-labs/agentcn (457 stars, last pushed 16d ago), licensed MIT. It adds 55 tokens to every session and 1,786 once invoked, about $0.0003 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 2026-08-30.

Related

Other skills, from other repositories

json-render-catalog

Skill "json-render-catalog" from yonatangross/orchestkit, covering json-render component catalogs, storybook → catalog import (#1529, 2026-04), new in 2026-04 → 2026-08 (json-render 0.14 → 0.20), directives — @json-render/directives (0.19) and registration.

yonatangross/orchestkit · 99 tokens

component-search

Search 21st.dev component registry for production-ready React components. Finds components by natural language description, filters by framework and style system, returns ranked results with install instructions. Use when looking for UI components, finding alternatives to existing components, or sourcing design system…

yonatangross/orchestkit · 57 tokens

agent-readme

Write or deslopify a registry agent's README (registry/ /README.md). Use when authoring a new agent README, trimming an existing one, or cutting AI-generated README noise ("these READMEs are too long / generic / repetitive", "deslopify the readmes", "fix the README"). Not for agent/instructions.md, the runtime prompt.

elie222/atom-eve · 83 tokens

revenue-pulse

Use when writing the weekly Stripe revenue & churn pulse — how to pull facts with the stripe and posthog-cli CLIs in the sandbox, then post to Slack.

elie222/atom-eve · 33 tokens

author-atom-eve-agent

Use when authoring or contributing a NEW Atom Eve agent to the registry (not installing an existing one) - designing what it does, wiring its capability and connections, and writing the README, SETUP.md, and agent/ source. Triggers include "build a new Atom Eve agent", "contribute an agent", "add a new agent to the…

elie222/atom-eve · 102 tokens

ppc-assist

Use when running PPC Assist Amazon PPC triage, interpreting PPC Assist metrics, reviewing search terms and suggestions, or preparing approval-gated account changes.

elie222/atom-eve · 34 tokens