Scout-ready infrastructure for agents

A boy scout with binoculars for your agent stack.

agent-look starts with a visible identity: a practical little scout that spots your screenshots, reads the trail, and brings back the useful part. The bigger idea is a family of tools, skills, and plugins that each know one job cold, then fade into the background as agents start doing the job directly or learn exactly when to use the tool for you.

Practical No ceremony. The tool does one real task with as little friction as possible.
Distinct Every tool has its own identity so agents and humans can learn what job it owns.
Invisible The best outcome is that the tool disappears because the workflow is now built in.

Each practical tool gets a sharp identity up front.

We are not building a pile of vague abstractions. We are building a family of practical tools that agents can trust. Each one is legible at the beginning, useful on day one, and progressively less visible as the agent stack absorbs the pattern.

agent-look

Finds the screenshots you just took, reads the text via Spotlight OCR, renames generic files, and reports back with the useful evidence. It is the scout with binoculars.

Spotting

Agent skills

Package repeatable moves into named capabilities. A good skill starts visible enough to teach the pattern, then becomes something the agent reaches for automatically at the right moment.

Learned routines

Agent plugins

Extend the agent into real tools and environments. Browser control, local state, external systems, file access: practical reach with a clear job boundary and a clean handoff back to the agent.

Reach
The point is not to keep the tool in the foreground forever. The point is to give it enough shape that agents and humans both know what it is for. Once that is learned, the tool should feel like infrastructure, not ceremony.
1

Name the job clearly.

Give the tool a crisp purpose and an identity strong enough that people understand when it should be used.

2

Make it frictionless.

One command, one habit, one obvious move. The tool should solve a real annoyance immediately, not after a migration project.

3

Let it disappear.

As the agent stack matures, the skill becomes implicit. The capability remains, but the user no longer has to think about the machinery.

Why agent-look works in the field instead of just in demos.

The scout metaphor is only useful if the tool is dependable. These implementation choices keep it light, practical, and boring in the right way.

It finds screenshots where shell tools fail.

It uses mdfind -onlyin … kMDItemIsScreenCapture=1 so iCloud-backed screenshot folders still work even when a direct filesystem walk hits macOS permission walls.

Spotlight metadata

It keeps tokens for when they matter.

Spotlight OCR via kMDItemTextContent handles most screenshots for free. Vision fallback is reserved for the cases where OCR is empty, so the tool stays fast and cheap.

Token discipline

It renames files without shell edge-case nonsense.

macOS screenshot names can include tricky spacing characters. fs.renameSync() with exact Spotlight paths avoids the fragile shell quoting path and just renames the file correctly.

Practical reliability

It already thinks like family infrastructure.

agent-look register wires into Claude Code, Claude Desktop, Gemini CLI, and Codex. That keeps the tool recognizable now while positioning it to become part of the default agent workflow later.

Multi-agent ready

Pack the scout once, then forget about it.

The install story should match the product story: direct, practical, no ceremony. Homebrew gets the tool onto the machine, then agent-look register wires it into the agent surfaces you already use.

Homebrew brew tap ashrocket/agent-look brew install agent-look agent-look register
Claude Code plugin /plugin marketplace add ashrocket/agent-look /plugin install agent-look@agent-look
Manual git clone git@github.com:ashrocket/agent-look.git cd agent-look && bash install.sh
Status and cleanup

Use agent-look status to inspect registrations and agent-look unregister to remove them cleanly.

What gets installed

The package includes the MCP server, the screenshot examiner agent, and the skill wiring so the scout is available in the tools that support it.

When something is off

File an issue if registration or screenshot discovery breaks. The goal is for this to feel invisible, so friction is a bug.

The family can keep growing without getting noisy.

`agent-look` is one member of a broader pattern. `agent-pb` handles the last useful artifact like a clipboard knife. Browser plugins like `superpowers-chrome` extend the same practical philosophy into live inspection and control. Different jobs, same standard: strong identity now, invisible usefulness later.

agent-pb Copies the last useful artifact without the selection dance.
superpowers-chrome Gives the agent hands and eyes in the browser when screenshots are not enough.
agent-look on GitHub The scout for screenshots, OCR, and lightweight evidence gathering.