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 skills add alexmx/peek --skill use-peekgit clone --depth 1 https://github.com/alexmx/peekWrote 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/skills/alexmx/peek/use-peek)<a href="https://agentmods.dev/skills/alexmx/peek/use-peek"><img src="https://agentmods.dev/badge/skills/alexmx/peek/use-peek/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/skills/alexmx/peek/use-peek"><img src="https://agentmods.dev/badge/skills/alexmx/peek/use-peek.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00051 | $0.03154 |
| Opus 5.5 | $0.00020 | $0.01262 |
| Sonnet 5.5 | $0.00010 | $0.00631 |
| Haiku 4.5 | $0.00005 | $0.00315 |
Grade A, and why
use-peek 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 7d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
peek — macOS Window Inspector CLI
peek inspects and drives native macOS app windows via Accessibility: tree inspection, element search, UI automation, screenshots, app lifecycle, menu bar.
Requirements: macOS 15+, Accessibility permission (and Screen Recording for peek capture). Run peek doctor --prompt to grant.
Always pass --format toon for LLM workflows (30-50% fewer tokens than JSON).
Workflow
Discover → act → verify in as few calls as possible.
peek launch --name Calculator --wait-for-window # → returns pid + windowID + windowTitle
peek find --app Calculator --role Button --title 5 --limit 1
peek action --app Calculator --do Press --role Button --title 5 --verify diff
- Discover with
peek launch --wait-for-window(returns windowID) orpeek apps --app X. Skippeek appsafter launch — it's already in the result. - Inspect structure with
peek tree(unfamiliar UI) orpeek find(known labels, much cheaper —--limit 1for existence checks). - Act+verify in one call:
peek action --verify diffreturns just the delta — usually 10x smaller than--verify tree. - Shortcuts (⌘S, Esc, arrows, F-keys): use
peek key, not menu navigation. - Menu items:
peek menu --click "Save"for menu BAR;peek action --do Pressonly works for items in an already-open menu.
Window targeting
All UI commands accept one of: --app NAME (case-insensitive substring), --pid PID, or a positional <window-id>. --app is simplest. Doesn't apply to launch / quit (use --bundle-id / --name).
For coordinate ops (click/drag/scroll) with a target set, peek verifies the point lands on that app/window — it raises an occluded target window, or errors (x, y) is over '<OtherApp>' if the coordinates are stale or over another window. On that error, re-read frames with peek find (windows move/resize) or pass a specific <window-id> for multi-window apps.
Window-less apps (Dock, Control Center, status-menu helpers): --app NAME or --pid scopes find/tree/action to the AXApplication root. Not in plain peek apps — go straight to peek find --app Dock --role AXDockItem (peek apps --include-panels does list the ones owning panels/status items).
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.
- 7d ago Changed 7d8f6557a050
- 1mo ago Changed · +5 lines 0bdfcd50c72c
- 1mo ago First seen · 155 lines · 51 tokens per session scan A 7c950419ab52
use-peek is a skill published in the GitHub repository alexmx/peek (5 stars, last pushed 8d ago), licensed MIT. It adds 51 tokens to every session and 3,154 once invoked, about $0.0002 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 skills, from other repositories
opencli-autofix
Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
surf
Control Chrome browser via CLI for testing, automation, and debugging. Use when the user needs browser automation, screenshots, form filling, page inspection, network/CPU emulation, DevTools streaming, or AI queries via ChatGPT/Gemini/Perplexity/Grok/AI Studio.
tauri-mcp-cli
Use the Tauri MCP CLI to start and recover driver sessions, automate Tauri webviews, capture UI state, debug IPC, and work with mobile or remote devices. Use whenever an agent needs to operate a Tauri v2 app from terminal commands.
acrawl-mcp
Drive the acrawl MCP server for browser automation, web scraping, and live DevTools inspection — navigating pages, clicking, filling forms, extracting structured data, and auditing network traffic, console errors, cookies, storage, performance, accessibility, and code coverage. Use this whenever a task touches acrawl…
agent-computer-use
REQUIRED for any task that involves operating a desktop application — opening apps, clicking buttons, typing into fields, pressing keys, scrolling, dragging, reading what's on screen, moving or resizing windows, or verifying state after an action. Always use the agent-cu CLI commands (open, snapshot, click, type, key…
debug-failed-action
When your last vsact returned an unexpected result (NOTFOUND, no visible state change, suspected silent failure), run this skill with the action's pre-token. Pulls console errors, network failures, and current performance into one report.