Borrowing it
Nothing to install: this file belongs to Ezra-Black/xcautokit. 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/Ezra-Black/xcautokit/main/AGENTS.mdgit clone --depth 1 https://github.com/Ezra-Black/xcautokitWrote 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/ezra-black/xcautokit/agents-md)<a href="https://agentmods.dev/instructions/ezra-black/xcautokit/agents-md"><img src="https://agentmods.dev/badge/instructions/ezra-black/xcautokit/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/ezra-black/xcautokit/agents-md"><img src="https://agentmods.dev/badge/instructions/ezra-black/xcautokit/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 | $0.01291 | $0.01291 |
| Opus 5.5 | $0.00516 | $0.00516 |
| Sonnet 5.5 | $0.00258 | $0.00258 |
| Haiku 4.5 | $0.00129 | $0.00129 |
Grade A, and why
xcautokit 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 23d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
XCAutokit
XCAutokit is the first-party MCP server for Xcode + iOS Simulator work in this environment. Prefer XCAutokit tools over any third-party Xcode/simulator MCP.
Setup
{
"mcpServers": {
"xcautokit": {
"command": "npx",
"args": ["-y", "xcautokit@latest", "mcp"]
}
}
}
Local checkout:
npm install && npm run build:go && npm link- MCP:
"command": "xcautokit", "args": ["mcp"](or.cursor/mcp.local.json) - For live Xcode IDE tools:
- Install Xcode 26.3+
- Open your project/workspace in Xcode
- Xcode → Settings → Intelligence → Allow external agents to use Xcode tools
- Optional env:
XCAUTOKIT_XCODE_BACKEND=auto|on|offMCP_XCODE_PIDXCAUTOKIT_WORKFLOWS=core(ordevice,ui,input,build,…) to shrink the tool listXCAUTOKIT_INTERRUPT_GUARD=offto disable input blocking on alerts
XCAutokit uses xcrun mcpbridge internally when available.
Host-owned intelligence
XCAutokit never runs agents or calls models. The connected host plans and may delegate to its own agents. Use one driver per simulator; let other agents review code, logs, and evidence or give them separate devices.
- Select an explicit simulator UUID when multiple devices are booted. Ambiguous targets fail.
device_claimreserves a simulator across calls; pass itsleaseTokento mutations and renew before expiry.device_releasefinishes the claim. Without a claim, each mutation takes a temporary lock. Claims are process-local; cross-process locks block other writers.ui_actresolves one fresh, enabled target, taps once, optionally checkswaitFor, and returns a compact observation.ui_waitpolls a stated condition within a deadline. A failed verification afterperformed:trueis not permission to repeat a tap.workflow_runexecutes 1–30 explicit steps, stops at the first failure, and persists evidence. Only successful runs ending in an explicit condition can be saved withworkflow_save. Replays make no decisions and never accept permissions.- Session defaults are isolated in memory. Explicit
XCAUTOKIT_SESSION_FILEopts into persistence.XCAUTOKIT_STATE_DIRselects the shared lock/evidence/workflow directory. build_run_simbuilds, installs the matching artifact, then launches on the exact selected simulator. Live Xcode tools are separate and follow the IDE's own destination; do not parallelize their mutations with a simulator driver.
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.
- 23d ago Changed · +16 lines · +389 tokens per session 6588df11c131
- 1mo ago Changed · +9 lines · +108 tokens per session 9323400699fe
- 2mo ago First seen · 74 lines · 794 tokens per session scan A 9f13104f9587
xcautokit AGENTS.md is an instructions file published in the GitHub repository Ezra-Black/xcautokit (1 stars, last pushed 24d ago), licensed MIT. It adds 1,291 tokens to every session, about $0.0052 per session on Opus 5.5. 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
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).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.