see

see is a command for coding agents from openclaw/Peekaboo. It costs 0 tokens per session (4,270 once invoked), scanned A, original, MIT.

A command that captures the current macOS interface with Peekaboo and extracts its accessibility information. It can also save screenshots annotated with element identifiers, positions, labels, and snapshot references.

In plain words
What is it for?
Use it to inspect a macOS screen, locate controls for automation, retrieve accessibility-tree details, or save an annotated screenshot for later reference.
Why use it?
It provides a current map of visible interface elements, so automation can target the right controls without guessing. The observation itself does not change which app or window has focus.

Command

Install

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.

agentmods
npx agentmods add commands/openclaw/peekaboo/see
Clone the repo
git clone --depth 1 https://github.com/openclaw/Peekaboo

Wrote 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.

agentmods badge for see

README.md
[![agentmods](https://agentmods.dev/badge/commands/openclaw/peekaboo/see.svg)](https://agentmods.dev/commands/openclaw/peekaboo/see)
Your own site
<a href="https://agentmods.dev/commands/openclaw/peekaboo/see"><img src="https://agentmods.dev/badge/commands/openclaw/peekaboo/see.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,270 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.04270
Opus 5 $0.00000 $0.02135
Sonnet 5 $0.00000 $0.00854
Haiku 4.5 $0.00000 $0.00427

Measured 3d ago against content hash 6e000ceeb1a4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

see 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 3d 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.

Origin

Copies of this mod

2 near-identical copies found in the catalogue:

  • doctor.ko — 86% identical, 184 lines differ
  • doctor.ja — 83% identical, 184 lines differ
docs/commands/see.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

peekaboo see

peekaboo see captures the current macOS UI, extracts accessibility metadata, and (optionally) saves annotated screenshots. CLI and agent flows rely on these UI maps to find fresh element IDs, bounds, labels, and snapshot IDs.

Observation is read-only with respect to focus: targeting a background app does not activate it or move its windows. With implicit host discovery and the standard daemon path, see prefers the current CLI build's deterministic build-scoped daemon and may auto-start it before considering a healthy Peekaboo.app host. This applies to pixel and AX-tree-only forms because their snapshots and capability decisions are host-memory state. An explicit --bridge-socket, a custom daemon socket, or --no-remote remains authoritative.

Every reusable snapshot is identified by a producer-generated reference with the exact form ps1_ followed by 32 lowercase ASCII hexadecimal digits, for example ps1_0123456789abcdef0123456789abcdef. The 32-digit suffix contains 128 random bits. The selected local or Bridge producer reserves that reference before storing detection results or screenshots; later store operations can update only a reference that producer already owns and cannot create one.

A later command that supplies a concrete reference finds its unique live authenticated producer before applying the normal daemon/app host preferences. Explicit routing remains a hard boundary: --bridge-socket probes only that host, and --no-remote checks only the local process. Either form refuses before action when its selected host does not own the reference. Legacy timestamp IDs are never accepted as action references; clean recognizes only their strict on-disk shape for cache removal.

# Capture frontmost window, print JSON, and save an annotated PNG
peekaboo see --json --annotate --path /tmp/see.png

# Target a specific app or window title
peekaboo see --app "Google Chrome" --window-title "Login" --json --path /tmp/chrome-login.png

# Crop one exact window without activating it
peekaboo see --window-id 12345 --roi 100,80,500,300 --json --path /tmp/window-roi.png

# Add host-local Vision text when an app exposes a sparse or incomplete AX tree
peekaboo see --app Calendar --window-id 12345 --ocr --json --path /tmp/calendar.png

Read the full file on GitHub · 169 lines

Changes

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.

  1. 3d ago First seen · 169 lines · 0 tokens per session scan A 6e000ceeb1a4

Subscribe to this mod's changes

see is a command published in the GitHub repository openclaw/Peekaboo (5,098 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 4,270 tokens. 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.