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/alrightryanx/claude-shadow/claude-mdgit clone --depth 1 https://github.com/alrightryanx/claude-shadowWrote 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/alrightryanx/claude-shadow/claude-md)<a href="https://agentmods.dev/instructions/alrightryanx/claude-shadow/claude-md"><img src="https://agentmods.dev/badge/instructions/alrightryanx/claude-shadow/claude-md.svg" alt="Measured on agentmods" 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 | $0.02332 | $0.02332 |
| Opus 5 | $0.01166 | $0.01166 |
| Sonnet 5 | $0.00466 | $0.00466 |
| Haiku 4.5 | $0.00233 | $0.00233 |
Grade A, and why
claude-shadow 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 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.
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 — 262 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code Instructions
When working on this codebase:
Project Overview
ShadowAI is an ecosystem of tools for voice-first AI assistance:
- Android app with Android Auto, Wear OS, and Google TV support
- PC bridge for SSH key exchange, clipboard sync, and Claude Code relay
- Claude Code plugin for mobile notifications and approvals
Project Tree
C:\shadow\
├── shadow-android/ # Main Android app
│ ├── app/ # Phone + Android Auto app
│ ├── shared/ # Shared library (backends, LLM)
│ │ └── src/main/cpp/
│ │ ├── llama.cpp/ # On-device LLM (git submodule)
│ │ └── whisper.cpp/ # On-device speech (git submodule)
│ └── wear/ # Wear OS app
│
├── shadow-bridge/ # PC companion app (Python)
│ └── shadow_bridge_gui.py # Main entry point
│
└── claude-shadow/ # Claude Code plugin
├── .claude-plugin/
│ ├── plugin.json # Plugin manifest
│ └── marketplace.json # Self-hosted marketplace
├── hooks/ # Claude Code hooks
└── scripts/ # PowerShell notification scripts
Component Descriptions
shadow-android (ShadowAI App)
Voice-first AI assistant for Android with multiple backends:
- SSH backends: Claude Code, Gemini CLI, Codex
- Cloud APIs: Anthropic, OpenAI, Gemini
- On-device: llama.cpp for offline LLM inference
- Features: Projects, Notes, Automations, DevOps integrations (Slack, Jira, PagerDuty)
shadow-bridge (ShadowBridge)
Windows PC companion that runs in system tray:
- Port 19284: Data receiver (clipboard, SSH keys)
- Port 19285: Network discovery broadcast
- Port 19286: Claude Code Companion relay
claude-shadow (Claude Code Plugin)
Plugin that sends notifications to ShadowAI when Claude Code needs input:
- Permission request notifications
- Session start/stop events
- Mobile approval/denial responses
- Reply via clipboard sync
Build Commands
Android App
cd shadow-android
./gradlew assembleDebug # Debug build
./gradlew assembleRelease # Release build
./gradlew test # Run tests
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 · 262 lines · 2,332 tokens per session scan A 6d38f2b59ae9
claude-shadow CLAUDE.md is an instructions file published in the GitHub repository alrightryanx/claude-shadow (9 stars, last pushed 8mo ago), licensed MIT. It adds 2,332 tokens to every session, about $0.0117 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 2026-08-31.
Other instructions, from other repositories
MobiAI-Core CLAUDE.md
Claude Code instructions for ArisGuimera/MobiAI-Core, covering mobiai — mobile development ai ecosystem, bootstrap and principles.
freetube CLAUDE.md
Instructions for leshkodev/freetube, covering claude.md, 1. what this project is, 2. non-negotiable constraints, 3. tech stack (locked) and 4. project structure.
ConsultMe CLAUDE.md
Claude Code instructions for Tarek-Bohdima/ConsultMe, covering claude.md, project context, common commands, module graph and conventions enforced by tooling.
phone-mcp CLAUDE.md
Claude Code instructions for premex-ab/phone-mcp, covering claude.md, project overview, build commands, architecture and module layout.
dev-challenge CLAUDE.md
Claude Code instructions for micheltlutz/dev-challenge, covering claude.md, the one rule to internalise, slash commands, subagents and skills.
android-media-pack CLAUDE.md
Instructions for sunnat629/android-media-pack, covering claude.md, what this repository is, commands, markdown lint (config in .markdownlint-cli2.jsonc) and shell script lint.