Borrowing it
Nothing to install: this file belongs to puritysb/AgentDeck. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/puritysb/AgentDeck/master/AGENTS.mdgit clone --depth 1 https://github.com/puritysb/AgentDeckWrote 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/puritysb/agentdeck/agents-md)<a href="https://agentmods.dev/instructions/puritysb/agentdeck/agents-md"><img src="https://agentmods.dev/badge/instructions/puritysb/agentdeck/agents-md/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.
<a href="https://agentmods.dev/instructions/puritysb/agentdeck/agents-md"><img src="https://agentmods.dev/badge/instructions/puritysb/agentdeck/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.07229 | $0.07229 |
| Opus 5.5 | $0.02892 | $0.02892 |
| Sonnet 5.5 | $0.01446 | $0.01446 |
| Haiku 4.5 | $0.00723 | $0.00723 |
Grade B, and why
AgentDeck AGENTS.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 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **hooks/** — installers for the Claude Code hooks (user-global `~/.claude/settings.json`; `settings.local.json` is dead at user scope and only cleaned up), the Codex lifecycle hooks (`~/.codex/config.toml`) and the Ope How it starts
The opening of the file, as written. The whole thing — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AgentDeck
Stream Deck+ controller for AI coding agents — a bidirectional local control system.
How this file is organised
This is the repository's only root instruction file; there is no CLAUDE.md. Claude Code (2.1.277+) loads
AGENTS.md whenever no CLAUDE.md, .claude/CLAUDE.md or CLAUDE.local.md exists in the working directory or
any directory above it — never add one of those here, it would silently replace this file. Codex injects the
root→cwd AGENTS.md chain and stops adding files once the combined size reaches project_doc_max_bytes
(32 KiB default), so this file plus esp32/AGENTS.md must stay under that together (pnpm docs:check gates both the
budget and any shadowing CLAUDE.md); OpenCode and Antigravity
read it by convention. It is the map; the rule bodies live in .claude/rules/*.md (tracked in git). Each rule file
declares paths: globs; Claude Code loads it automatically when a file matching those globs is touched. Codex,
OpenCode and Antigravity have no path-conditional loading and must read the matching file before the first edit
in that area (.codex/rules is exec policy, not instructions). Measurements behind a rule are in the topic doc it
links; incidents are in docs/devlog/entries/ (grep them; see DEVELOPMENT_LOG.md).
| Rule file | Loads when touching | Covers |
|---|---|---|
| apme-eval | bridge/src/apme/**, shared/src/sample.ts, Swift Daemon/Apme/** |
SessionSample SSOT, task/turn segmentation, end_source, judge chain mlx → foundationModels, transport gate, JSON parser, backlog drain, task titles, per-turn cost, reference judge |
| openclaw-gateway | bridge/src/openclaw*, shared/src/gateway-*, Swift Daemon/Gateway/** |
sessions.subscribe, per-key runs, transcript feed, link instability, three-answer health frames |
| observed-sessions | bridge/src/hook-*, kiro-*, codex-*, claude-*, hooks/**, Swift Daemon/Session/** |
Hook format + installers, per-agent observation, PERM sources and the Claude hold predictor, subagent census, three identity axes, two id forms |
| usage-quota | bridge/src/usage-*, codex-rate-limits*, Swift UsageAPIClient/UsageRelayFreshness |
Codex window axes (stale / aged / plan / limit family), cache TTL vs poll, Claude quota recovery |
| esp32-flash | esp32/**, tools/web-flasher/**, bridge/src/esp32*, shared/src/esp32-boards.ts |
Board map, merged factory image, preflight refusal, post-write reset, serial-suspend lease, serial teardown, mDNS before radio-off |
| daemon-lifecycle | bridge/src/daemon*, auth.ts, pairing-*, network-posture.ts, Swift DaemonService/Server/** |
Timeline persistence, network posture, LAN default-deny, pairing, token custody, ownership, mDNS name, supervisor routing, port intent, build identity, restart verification, wake detector |
| swift-daemon | apple/** |
@DaemonActor, NSApp.windows, scheduled self-restart, epoch-guarded teardown, DaemonOwnershipChange, ObjC exceptions, bounded awaits, progress indicators |
| devices-and-wire | shared/**, plugin*/**, android/**, bridge/src/pixoo/**, modules/**, Swift Daemon/Modules/** |
Surface boundary, weather cache, wire semantics, unknown agentType, back-dated dedup, device-keyed work, baked animation loops, dot-matrix masks, action naming, Ulanzi WASM packaging, BLE worker breaker |
| managed-sessions | bridge/src/cli.ts, pty-manager.ts, adapters/** |
--weight, AGENTDECK_*_ARGS, node-pty helper repair |
| design-system | design/**, docs/**, UI dirs, *.css, *.html |
Seven design rules, token mirrors, coverage and Markdown gates |
| apple-release | apple/**, RELEASING.md, .github/workflows/** |
App Store invariants, five release states, CHANGELOG-rendered release bodies |
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.
- 2d ago Changed · +2 lines · +33 tokens per session 1a933ec7607c
- 9d ago Changed · +47 tokens per session c0ea3c7fa7e5
- 12d ago Changed · -63 tokens per session 9d5aa4cb5b92
- 15d ago Changed · -11 tokens per session e0ba0663df97
- 20d ago Changed · +8 tokens per session e3551e24bc88
- 29d ago Changed · +187 lines · +6,577 tokens per session scan A → B 420ca1810429
- 1mo ago Changed · -37 lines · -1,722 tokens per session 9337c184d314
- 1mo ago Changed · +19 lines · +840 tokens per session scan B → A e6afa0343546
- 1mo ago First seen · 40 lines · 1,520 tokens per session scan B 7115b1211ea5
AgentDeck AGENTS.md is an instructions file published in the GitHub repository puritysb/AgentDeck (274 stars, last pushed today), licensed MIT. It adds 7,229 tokens to every session, about $0.0289 per session on Opus 5.5. 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 2026-08-30.
Other instructions, from other repositories
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.