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/appclawhq/appclaw/claude-mdgit clone --depth 1 https://github.com/appclawhq/AppClawWhat 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 | $0.02529 | $0.02529 |
| Opus 5 | $0.01264 | $0.01264 |
| Sonnet 5 | $0.00506 | $0.00506 |
| Haiku 4.5 | $0.00253 | $0.00253 |
Grade A, and why
AppClaw 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 yesterday.
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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What is AppClaw?
AppClaw is an agentic AI layer for mobile automation (Android & iOS). Users describe goals in plain English and AppClaw orchestrates device interactions through appium-mcp (Model Context Protocol). It supports multiple LLM providers (Anthropic, OpenAI, Google Gemini, Groq, Ollama) via the Vercel AI SDK.
Build & Run Commands
npm run build # TypeScript compilation (tsc) → dist/
npm run typecheck # Type-check without emitting
npm start # Run via tsx (dev mode)
npm start "goal" # Run with a goal argument
npm run dev # Run with file watching
npx tsx tests/verify-parsing.ts # Run parsing verification tests
No formal test framework (Jest/Vitest) is configured. Tests are ad-hoc scripts in tests/.
Architecture
Entry Point & CLI Modes (src/index.ts)
The CLI routes to 6 modes based on flags:
- Interactive (default) — prompts for platform/device/goal, runs agent loop
- YAML Flow (
--flow file.yaml) — declarative automation, zero LLM cost - Terminal Studio (
--tui, or its alias--playground) — multi-screen Ink app: platform/device picker, slash-command palette, step recording,/goalagent runs, settings, run history, and device mirroring —/streamrenders the screen inside the terminal (see Terminal Studio below). This is the interactive mode; the old--playgroundREPL was removed and the flag now routes here. - Explorer (
--explore) — PRD → YAML test flow generation - Record/Replay (
--record,--replay) — capture and replay sessions - Report (
--report) — Express server serving HTML reports from.appclaw/runs/
The interactive and goal-direct paths also accept --export [path] (optionally --export-dir <dir>) to write a replayable @appclaw/runner spec when the goal completes. Path resolution: empty → EXPORT_DIR/<goal-slug>.test.ts (EXPORT_DIR defaults to tests, the runner's own testDir, so an export is runnable where it lands); bare filename → EXPORT_DIR/<name>; anything with a directory hint → used verbatim. Implementation in src/sdk/goal-export.ts (translator + renderer) called from src/index.ts after the agent loop.
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.
- yesterday First seen · 96 lines · 2,529 tokens per session scan A f570dfeddfb8
AppClaw CLAUDE.md is an instructions file published in the GitHub repository appclawhq/AppClaw (109 stars, last pushed 5d ago), licensed Apache-2.0. It adds 2,529 tokens to every session, about $0.0126 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
Maestro AGENTS.md
Instructions for mobile-dev-inc/Maestro, covering agents.md — maestro, module map, e2e test fixtures (e2e/), passing/ vs failing/ suites and test-e2e.yaml workflow contract.
cindy AGENTS.md
Instructions for makecindy/cindy, covering cindy 客户端仓：agent 工作入口, 仓库边界, 规则组织, 当前规则索引 and 通用工作流程.
serve-sim AGENTS.md
Instructions for EvanBacon/serve-sim, covering native build notes, e2e testing with agent-browser and e2e testing via the serve-sim cli.
maui-labs AGENTS.md
Instructions for dotnet/maui-labs, covering agent instructions, repository overview, products, technology stack and building.
Xcode-Build-Optimization-Agent-Skill copilot-instructions.md
Instructions for AvdLee/Xcode-Build-Optimization-Agent-Skill, covering repository mission, agent skill requirements, behavior constraints, build optimization focus areas and common issues to flag.
react-native-restart AGENTS.md
Instructions for avishayil/react-native-restart: This project keeps its agent/AI onboarding guidance in a single canonical file.