What is agentsd CLAUDE.md?

Project instructions for agentsd, a Stream Deck plugin that coordinates local AI-agent sessions through Claude Code HTTP hooks. They describe its npm build commands, architecture, and session state machine.

Use them when building, testing, linking, debugging, or changing agentsd, especially work involving SessionManager, HTTP hooks, Stream Deck controls, or session transitions.

They give coding agents the repository's required commands and explain how hook requests become Stream Deck updates, including permission and elicitation states.

agentsd CLAUDE.md — Instructions file for coding agents by paultyng: Project instructions for agentsd, a Stream Deck plugin that…

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/paultyng/agentsd/claude-md

Clone the repo

git clone --depth 1 https://github.com/paultyng/agentsd

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/paultyng/agentsd/claude-md.svg)](https://agentmods.dev/instructions/paultyng/agentsd/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/paultyng/agentsd/claude-md"><img src="https://agentmods.dev/badge/instructions/paultyng/agentsd/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 447 This file is loaded in full into every session.

When invoked 447 The same file — it is already loaded in full.

Security scan B 1 finding. 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.

Model	Per session	Once invoked
Fable 5	$0.00447	$0.00447
Opus 5	$0.00224	$0.00224
Sonnet 5	$0.00089	$0.00089
Haiku 4.5	$0.00045	$0.00045

Measured 4d ago against content hash b8f0439f0add, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

agentsd CLAUDE.md scanned grade B with 1 finding 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- `npm run hooks:install` / `npm run hooks:uninstall` — manage Claude Code HTTP hooks in `~/.claude/settings.json`

CLAUDE.md · 37 lines

What it actually says

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project

agentsd — a Stream Deck plugin for orchestrating local AI agent usage. Early stages.

Build & Test

Use npm scripts only — no Taskfile or Makefile.

npm run build — rollup build → com.paultyng.agentsd.sdPlugin/bin/plugin.js
npm run watch — rollup watch mode
npm run dev — Stream Deck dev mode
npm run link / npm run unlink — link/unlink plugin in Stream Deck
npm run hooks:install / npm run hooks:uninstall — manage Claude Code HTTP hooks in ~/.claude/settings.json
npm run debug:hooks — interactive hook debug script

Architecture

Claude Code HTTP hooks → Plugin HTTP server (127.0.0.1:9200) → SessionManager → Stream Deck button/dial updates. No bridge daemon, no PTY parsing. PermissionRequest hooks hold HTTP response open (120s timeout) for approve/deny from hardware buttons.

State machine

Sessions move through 5 states: DISCONNECTED → IDLE → PROCESSING → AWAITING_PERMISSION / AWAITING_ELICITATION. Auto-created sessions (missed SessionStart) start as IDLE. Current-state guards prevent nonsensical transitions (e.g. tool events from DISCONNECTED). PostToolUse intentionally stays in PROCESSING — Stop moves to IDLE.

Key behaviors

Auto-foreground: Elicitation and PermissionRequest events bring the session to the active slot.
Stale pruning: Sessions with no activity for 60s are pruned (PRUNE_INTERVAL_MS = 60s check).
Permission timeout: 120s (PERMISSION_TIMEOUT_MS). Sends explicit deny on timeout.
Action DI pattern: A single setManager() call in plugin.ts wires the SessionManager into all ManagedAction subclasses via a shared module-level reference in actions/base.ts.

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.

4d ago First seen · 37 lines · 447 tokens per session scan B b8f0439f0add

Subscribe to this mod's changes

agentsd CLAUDE.md is an instructions file published in the GitHub repository paultyng/agentsd (10 stars, last pushed 22d ago), licensed MIT. It adds 447 tokens to every session, about $0.0022 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

ai-agents claude-code stream-deck-plugin streamdeck

OpenDeck AGENTS.md

AGENTS.md instructions for nekename/OpenDeck, covering opendeck development guide, token efficiency, architecture overview, core architecture pattern and project structure.

nekename/OpenDeck · 1,955 tokens

streamdeck-mcp CLAUDE.md

Instructions for verygoodplugins/streamdeck-mcp, covering claude.md, development commands, setup, run server (for claude desktop) and lint.

verygoodplugins/streamdeck-mcp · 826 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

agentsd CLAUDE.md