Skip to content
Get packs

OpenCLI Browser Control

Lets an agent drive Chrome to read pages, fill forms, capture API traffic and manage tabs

Source Download skill
Repository
jackwener/OpenCLI
Stars
30,833
Licence
Apache-2.0
Files
1 file

What it does

Once installed, the agent can operate a real Chrome browser through the OpenCLI command-line tool: open pages, read their content and interactive elements, click, type and fill forms, take screenshots, and inspect network requests the page makes. It can work inside sessions that reuse your logged-in Chrome profile, and it can group, rename and move Chrome tabs and windows through Chrome's native APIs.

Great for

  • Developers who want an agent to test or inspect a web app in the browser
  • People automating repetitive tasks on sites they are already logged into
  • Analysts who need to discover which API endpoints a web page calls

Skip it if

  • Sensitive accounts such as banking or email, because the agent can act inside logged-in sessions and run arbitrary JavaScript with eval (risk of unintended changes)
  • Users who do not have Chrome or the opencli tool installed, since the skill only teaches commands and cannot run the browser itself

Try asking

  • “Open example.com in a research session, search for a product and read the results”
  • “Capture the API requests a page makes when I click Load More, and summarise the response shapes”
  • “Group my open Chrome tabs about the project into a group titled Research”
  • “Fill in the contact form on this page and verify the values before submitting”
Step-by-step guide: how to use OpenCLI Browser ControlInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    opencli-browser.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip opencli-browser.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip opencli-browser.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Use a dedicated session name per task so page work stays isolated, and run opencli doctor if the connection fails
  • Check the help output with --help -f json before guessing arguments, and treat static network candidates as unverified leads
  • Prefer bind only for tabs you intend the agent to navigate; use owned sessions for tasks that create their own tabs

What's inside

  • SKILL.md

Redistributed unmodified from jackwener/OpenCLI under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.