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/standardbeagle/agnt/agents-mdgit clone --depth 1 https://github.com/standardbeagle/agntWhat 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 | $0.05963 | $0.05963 |
| Opus 5 | $0.02982 | $0.02982 |
| Sonnet 5 | $0.01193 | $0.01193 |
| Haiku 4.5 | $0.00596 | $0.00596 |
Grade A, and why
agnt AGENTS.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.
How it starts
The opening of the file, as written. The whole thing — 271 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Claude Code 治此 repo 之綱：唯導向與不變式；詳見 docs/、.claude/rules/（見 Reference Map）。
Project Overview
agnt：賦 AI coding agents 以 browser superpowers；通 AI agent 與 browser，作 real-time debug、UI wireframe、visual feedback。
- Version: 0.15.4
- Repository: https://github.com/standardbeagle/agnt
Binaries:
agnt: 主 CLI（唯一實建 binary）agnt-daemon: daemon auto-start 副本（避 sandbox 禁 fork）devtool-mcp: 舊名 alias（backwards compat）
CLI Subcommands: mcp (MCP server), serve, run (PTY wrapper), daemon, session (hosts lists detachable sessions; kill terminates one), attach (detach/reattach a daemon-owned PTY), ssh (remote session-host client with reconnect and forwarding), push (SFTP delivery to an active remote session), init (setup-only, no relaunch), skills (install agnt skills via npx skills + register MCP), shim (shim watcher/install), monitor (event stream), errors, ci, doctor, notify, up, upgrade, completion, ai (interactive AI — Claude-only, stream-json), acp (any ACP agent via coder/acp-go-sdk: gemini/opencode/claude-code-acp; one-shot + overlay/cooked REPL; fs+terminal caps; deterministic alert gate), hook (telemetry forwarder), activate/license (Pro license activation + management — offline lk validation, see internal/license/)。setup-project is a skill (/agnt:setup-project), not a cobra command.
Core Architecture Decisions:
- Binary copies instead of self-exec：sandbox（如 Claude Code）禁 binary fork/exec self；別本 binary 可避之。
agnt runworkaround for MCP notifications：MCP servers 不能 push notifications；agnt run以 PTY 包 AI tools，注 browser events 為 synthetic stdin：Browser → Proxy → HTTP POST → Overlay (port 19191) → PTY stdin → AI Tool- System prompt/context delivery：起 AI agents 時自注或持久化 agnt context；Claude Code 用
--append-system-prompt；Gemini, Copilot, Aider 等 normal sessions 寫入 agent context file，setup mode 才用 stdin prompt。詳docs/agent-adapters.md。
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 First seen · 271 lines · 5,963 tokens per session scan A 085b22a21f1a
agnt AGENTS.md is an instructions file published in the GitHub repository standardbeagle/agnt (19 stars, last pushed 4d ago), licensed Apache-2.0. It adds 5,963 tokens to every session, about $0.0298 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.
Other instructions, from other repositories
paparazzi CLAUDE.md
Claude Code instructions for nicksenap/paparazzi, covering claude.md, project overview, commands, install & build and development (watch mode).
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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 oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).