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/ghostwright/ghost-os/claude-mdgit clone --depth 1 https://github.com/ghostwright/ghost-osWhat 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 | $0.01958 | $0.01958 |
| Opus 5 | $0.00979 | $0.00979 |
| Sonnet 5 | $0.00392 | $0.00392 |
| Haiku 4.5 | $0.00196 | $0.00196 |
Grade A, and why
ghost-os 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 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ghost OS -- Developer Guide
Accessibility-first perception and execution engine that gives AI agents the ability to see, understand, and operate any macOS application through the accessibility tree and visual perception.
Build & Run
swift build # Requires Swift 6.2+ (swiftly), macOS 14+
.build/debug/ghost mcp # Start MCP server (stdio)
.build/debug/ghost setup # Interactive first-run wizard
.build/debug/ghost doctor # Diagnostic health check
.build/debug/ghost status # Quick status
.build/debug/ghost version # Version info
Release: scripts/build-release.sh builds tarball, verifies 3-way version
consistency (Types.swift + server.py + ghost-vision), and codesigns the binary.
Architecture
MCP-first, single-threaded synchronous. No persistent daemon, no socket IPC, no async state model. Every MCP tool call queries the AX tree fresh.
AI Agent (Claude Code / Cursor / any MCP client)
|
| MCP Protocol (stdio, auto-detects Content-Length vs NDJSON)
|
ghost mcp (Swift binary, @MainActor)
|
+-- Perception (7 tools) -- AX tree via AXorcist
+-- Annotate (1 tool) -- Set-of-Marks labeled screenshots, zero ML
+-- Actions (10 tools) -- AX-native -> CDP -> VLM -> synthetic cascade
+-- Wait (1 tool) -- 6 polling condition types
+-- Recipes (5 tools) -- JSON workflow storage and execution
+-- Vision (2 tools) -- ShowUI-2B VLM via Python sidecar
+-- Learning (3 tools) -- CGEvent tap recording for recipe synthesis
|
+-- AXorcist (github.com/steipete/AXorcist, from: "0.1.0")
+-- ScreenCaptureKit (linked framework)
The learning subsystem is the one exception to single-threaded: CGEvent tap runs on a dedicated nonisolated background Thread with os_unfair_lock and CFRunLoop.
stdout is sacred. All MCP protocol data flows through stdout. A single stray
print() or Swift.print() will corrupt the protocol and break the connection.
All logging goes through Log (writes to stderr).
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 · 161 lines · 1,958 tokens per session scan A 2b1d48c315d9
ghost-os CLAUDE.md is an instructions file published in the GitHub repository ghostwright/ghost-os (1,651 stars, last pushed 5mo ago), licensed MIT. It adds 1,958 tokens to every session, about $0.0098 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
open-codex-computer-use AGENTS.md
AGENTS.md instructions for iFurySt/open-codex-computer-use, covering open-codex-computer-use, 每轮开始先读, 代码改完前要读, 按任务需要选读 and 工作规则.
open-codex-computer-use CLAUDE.md
Claude Code instructions for iFurySt/open-codex-computer-use, a project described as: 👾 Open Computer Use – Open-Source Alternative to Codex Computer Use.
dsh-computer-use AGENTS.md
Instructions for Anionex/dsh-computer-use, a project described as: 为 DeepSeek Harness 提供电脑控制插件：新鲜 Accessibility 观测、过期状态拒绝、作用域权限与安全输入（目前支持macos）｜Accessibility-first macOS Computer Use bundle for DSH with fresh observations, stale-state rejection, scoped permissions, and safe input.
DarwinRelay AGENTS.md
AGENTS.md instructions for dcierra/DarwinRelay, covering agents.md, what darwinrelay is, read these files first, repository map and canonical identities.
agent-computer-use-mcp AGENTS.md
AGENTS.md instructions for astraclawteam/agent-computer-use-mcp, covering agent worker rules, read order, repository purpose, project structure and branch and pr rules.
OpenAra AGENTS.md
Instructions for Aradotso/OpenAra, covering openara agent guide, read first, working rules and where things live.