New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Browser

by BlackBeltTechnology in BlackBeltTechnology/pi-agent-dashboard

Browser automation via the agent-browser CLI. An agent skill from BlackBeltTechnology/pi-agent-dashboard.

Apache-2.0Auto-check passedProductivity & Automation

Install Browser

skills CLI
$ npx skills add BlackBeltTechnology/pi-agent-dashboard --skill browser -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install BlackBeltTechnology/pi-agent-dashboard browser --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/BlackBeltTechnology/pi-agent-dashboard.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/extension/.pi/skills/browser .claude/skills/browser && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
browser
GitHub stars
315
Used in
1 other repo
Token cost
~2k tokens
SKILL.md length
839 words
Files
19 (incl. scripts, references)
Skills in repo
66
Repo updated
First seen
Licence
Apache-2.0

At a glance

Browser automation via the agent-browser CLI. An agent skill from BlackBeltTechnology/pi-agent-dashboard.

  • Works in 2 steps: references/dashboard-relay.md when the… → references/own-browser.md (legacy…
  • The user needs to drive websites
  • SKILL.md covers Step 0a — Preflight:…, Step 0b — Auto-detect: route…, Step 1 — Read the matched… and Notes
  • Runs Shell scripts from its folder; calls npm, curl and npx

What it does

Browser is an agent skill from BlackBeltTechnology/pi-agent-dashboard. Browser automation via the agent-browser CLI. Use when the user needs to drive websites or Electron apps — navigating, forms, clicks, screenshots, data extraction, testing web apps, UI checks, the dashboard Electron shell, or the user's own logged-in Chrome via the browser relay or Panerelay (SSO/2FA). Triggers: "open a website", "take a screenshot", "test this web app", "use my own browser".

Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 20 other files, including scripts and reference files (for example `SKILL.md.AGENTS.md`, `UPSTREAM.md` and `references/challenge.md`).

It sits in Productivity & Automation, covering Browser automation and Authentication. The repository describes itself as: Real-time web dashboard for pi coding-agent sessions. Multi-session view, live chat mirroring, integrated terminal, diff viewer, pi-flows execution, and mobile-first remote… The licence is Apache-2.0.

When your agent uses it

  • The user needs to drive websites
  • Electron apps — navigating
  • Data extraction
  • Testing web apps

Example prompts

  • “open a website”
  • “take a screenshot”
  • “test this web app”
  • “/browser”

Requirements

  • Node.js
  • A Bash shell
  • Pre-approved tools (allowed-tools): Bash(agent-browser:*), Bash(npx agent-browser:*), Bash(npx @panerelay/setup:*), Bash(curl:*)

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. references/dashboard-relay.md when the pi session runs on the
  2. references/own-browser.md (legacy Panerelay bridge) only when the

What it can do on your machine

Read from SKILL.md and the folder at commit e23e533. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Bash(agent-browser:*)
    • Bash(npx agent-browser:*)
    • Bash(npx @panerelay/setup:*)
    • Bash(curl:*)

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 3 files in scripts/ (Shell), which the agent can run.

    Shell commands in SKILL.md call:

    • npm
    • curl
    • npx

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npm, curl and npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Browser loads about 2k tokens when it runs, and up to ~40k if it reads all its reference files. Until then it costs about 101 tokens; SKILL.md has 839 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~101
When it runs · the whole SKILL.md, loaded when a task matches
~2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~40k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from BlackBeltTechnology/pi-agent-dashboard at commit e23e533, republished under its Apache-2.0 licence (© BlackBeltTechnology). 839 words, ~2,040 tokens.

Download SKILL.mdSave it as .claude/skills/browser/SKILL.md (or your agent's skills folder). This skill also uses 18 other files; get the full folder from GitHub.
name
browser
description
Browser automation via the `agent-browser` CLI. Use when the user needs to drive websites or Electron apps — navigating, forms, clicks, screenshots, data extraction, testing web apps, UI checks, the dashboard Electron shell, or the user's own logged-in Chrome via the browser relay or Panerelay (SSO/2FA). Triggers: "open a website", "take a screenshot", "test this web app", "use my own browser".
allowed-tools
Bash(agent-browser:*), Bash(npx agent-browser:*), Bash(npx @panerelay/setup:*), Bash(curl:*)
license
Apache-2.0
metadata.author
pi-dashboard
metadata.version
1.2
metadata.vendoredFrom
agent-browser
metadata.vendoredVersion
0.27.0
metadata.verifiedAgainstCli
0.33.2

browser

Composite skill that gives the agent eyes and hands for any browser-driven task — web pages or Electron desktop apps — via the agent-browser CLI.

Three recipes:

  • Web automation — generic web pages plus Pi Dashboard-specific helpers (dashboard URL detection, responsive testing, console-error hunting). Reference: references/web.md.
  • Electron automation — drive any Chromium-based Electron app, including a worked example for the Pi Dashboard's own shell via --debug-cdp. Reference: references/electron.md.
  • The user's own logged-in browser — reach SSO/2FA-gated sites using the user's real cookies. Two paths:
    • Pi Dashboard browser relay (preferred when the pi session runs on the dashboard host) — pair the Playwright Chrome Extension, connect to a loopback CDP URL. Reference: references/dashboard-relay.md.
    • Panerelay extension bridge (legacy) — a native-messaging host for when there is no loopback dashboard. Reference: references/own-browser.md.

Step 0a — Preflight: agent-browser CLI must be installed

The skill does not bundle the CLI. Verify it through the dashboard tool registry — recommend-only, nothing installs without an explicit --install:

bash
pi-dashboard-ensure <extension-package-root>/package.json

<extension-package-root> is the package that ships this skill — three levels above this SKILL.md's .pi/skills/browser/ directory (in a global install: $(npm root -g)/@blackbelt-technology/pi-dashboard/node_modules/@blackbelt-technology/pi-dashboard-extension). The manifest (pi.tools) declares agent-browser (resolve) and chromium (pw-browser); the registry answers present / recommended / blocked per tool.

  • agent-browser: present → continue to Step 0b.

  • agent-browser: recommended or blocked → halt and tell the user:

    The agent-browser CLI is not installed. Install it as a pi extension so the browser tool is registered in your pi session too:

    pi install npm:pi-agent-browser

    Then re-invoke the skill.

  • chromium missing → mention the registry's Install hint (npx playwright@1.62.1 install chromium) only when the chosen recipe needs a CDP browser; the dropdown on the Settings → Tools row carries it.

Do not attempt npm install, pi install, or any other install command on the user's behalf — they should make that choice explicitly.

Step 0b — Auto-detect: route to the right recipe

After the preflight, decide which recipe applies. Run this 2-line probe:

bash
if command -v lsof >/dev/null 2>&1; then
  CDP_LIVE=$(lsof -ti :9222 >/dev/null 2>&1 && echo yes || echo no)
else
  CDP_LIVE=$(nc -z 127.0.0.1 9222 2>/dev/null && echo yes || echo no)
fi
PD_RUNNING=$(pgrep -f "Pi Dashboard|pi-dashboard" >/dev/null 2>&1 && echo yes || echo no)
# Browser relay: the dashboard's own capability probe (see dashboard-relay.md).
RELAY=$(curl -fsS http://localhost:8000/api/browser/status 2>/dev/null || echo unavailable)
echo "CDP_LIVE=$CDP_LIVE PD_RUNNING=$PD_RUNNING RELAY=$RELAY"

Routing rule:

CDP_LIVEPD_RUNNINGRoute toWhy
yesyesreferences/electron.mdPi Dashboard Electron shell is attachable
noyesreferences/electron.mdPi Dashboard is up but CDP off — electron.md shows the --debug-cdp launch instruction
anynoreferences/web.mdNo Electron target running; default to web

Override: if the user's request is explicitly about a website (URL, HTTPS host, "the dashboard at localhost:8000", etc.), route to references/web.md regardless of what's running. Intent wins over capability.

Override: if the user's request is explicitly about an Electron app that isn't Pi Dashboard (Slack, VS Code, Figma, …), route to references/electron.md even when PD_RUNNING=no — that recipe covers launching any Electron app with --remote-debugging-port.

Override (login state): if the target needs the user's existing session — "use my browser", "I'm already logged in", SSO/2FA, an internal tool the agent cannot sign into — route to a login-state recipe regardless of the probe. Pick by capability, in this order:

  1. references/dashboard-relay.md when the pi session runs on the dashboard host and the probe's RELAY is a JSON object with "enabled":true and "canOpenChrome":true. This is the preferred path; it needs no extra install and the dashboard holds the pairing token.
  2. references/own-browser.md (legacy Panerelay bridge) only when the relay is unavailable (RELAY=unavailable) or the session is remote.

The default bundled browser starts logged out and cannot be made to inherit those logins; --profile copies the profile and still loses macOS Keychain-encrypted cookies. Prefer the bundled browser whenever login state is irrelevant — it is faster and more capable. Never fall back to it for a login-state task.

Show full SKILL.md (280 more words)Show less

Step 1 — Read the matched recipe and execute

Read the reference file selected above, then follow its workflow. All references are self-contained; you do not need to read more than one.

Challenge encountered: if at any point the page is an interactive anti-bot challenge (Cloudflare Turnstile, hCaptcha, reCAPTCHA, "Checking your browser…") that withholds the content you asked for, stop and follow references/challenge.md. Do not retry, do not switch --session, --profile, or provider, and do not run close --all.

Notes

  • Vendoring: references/web.md and references/electron.md are snapshots of upstream agent-browser skill content (core and electron) at CLI version 0.27.0. See UPSTREAM.md for refresh procedure and LICENSE for upstream attribution. references/own-browser.md, references/dashboard-relay.md, and references/challenge.md are authored, not vendored — an upstream refresh must not overwrite them.
  • Login-state recipes: references/dashboard-relay.md is the preferred path (Pi Dashboard browser relay); references/own-browser.md is the legacy Panerelay/native-messaging path for remote sessions with no loopback dashboard.
  • Surfaces share one browser: the browser MCP tool and Bash: agent-browser … drive the same daemon and same Chrome, with a shared active page. Probing with one mid-task can silently move the page out from under the other. Pick one surface per task; use --session <name> for genuinely isolated browsers.
  • Known wrapper bug: the MCP tool's eval echoes the JS source instead of executing it (and mangles regex escapes) — still present with CLI 0.33.2, so it is a pi-agent-browser wrapper defect, not a CLI one. Use Bash: agent-browser eval … instead.
  • No CLI bundled: agents installing the bridge extension get the skill text but not the CLI; install on demand per Step 0a.
  • User-local override: if the user's project has its own .pi/skills/browser/ skill, pi's local-wins precedence applies and this skill is shadowed — that's by design.

© BlackBeltTechnology, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 18 other files (scripts, references) in packages/extension/.pi/skills/browser of BlackBeltTechnology/pi-agent-dashboard.

  • SKILL.md
  • LICENSE
  • SKILL.md.AGENTS.md
  • UPSTREAM.md
  • references/challenge.md
  • references/challenge.md.AGENTS.md
  • references/dashboard-relay.md
  • references/dashboard-relay.md.AGENTS.md
  • references/electron.agent.md
  • references/electron.md
  • references/electron.md.AGENTS.md
  • references/own-browser.md
  • references/own-browser.md.AGENTS.md
  • references/web.agent.md
  • references/web.md
  • references/web.md.AGENTS.md
  • scripts/check-panerelay.sh
  • scripts/check-panerelay.sh.AGENTS.md
  • scripts/detect-dashboard.sh

Open the folder on GitHubat commit e23e533

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in BlackBeltTechnology/pi-agent-dashboard, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Browser next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Browser compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Browser this skillBlackBeltTechnology/pi-agent-dashboard3151 repos~2kAutomated safety check: PassApache-2.0
Zhihu Publishbifrost-proxy/bifrost160—~1.5kAutomated safety check: PassMIT
Agent BrowserHHU3637kr/skills145—~1.9kAutomated safety check: PassNone
Bitwarden Browser Aac Loginaiskillstore/marketplace430—~1.1kAutomated safety check: PassNone
Sap Browser Automationsecondsky/sap-skills462—~3.3kAutomated safety check: PassGPL-3.0
Xiaohongshu HubOpenMinis/MinisSkills444—~2.7kAutomated safety check: PassMIT

Similar skills

  • Zhihu Publish

    bifrost-proxy/bifrost

    Publish Markdown articles to Zhihu Columns through authentication context recovered locally from Bifrost traffic and a built-in publish payload schema, save drafts, prevent duplicate posts, and…

    160 GitHub stars~1.5k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed

  • Agent Browser

    HHU3637kr/skills

    Headless browser automation for AI agents using agent-browser CLI.

    145 GitHub stars~1.9k tokensUpdated 15 days ago
    Productivity & AutomationAuto-check passed

  • Bitwarden Browser Aac Login

    aiskillstore/marketplace

    Use Bitwarden Agent Access CLI (aac) with agent-controlled browser login flows.

    430 GitHub stars~1.1k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed

  • Sap Browser Automation

    secondsky/sap-skills

    A skill your agent uses when an agent must inspect or operate an authenticated SAP web UI through an in-app Browser, Microsoft Edge CDP, or an existing Playwright client, especially when SAP SSO…

    462 GitHub stars~3.3k tokensUpdated 3 days ago
    Productivity & AutomationAuto-check passed

  • Xiaohongshu Hub

    OpenMinis/MinisSkills

    Skill for reading and writing Xiaohongshu (XHS) data using Python + UV.

    444 GitHub stars~2.7k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed

  • Weibo Hub

    OpenMinis/MinisSkills

    A skill for reading and writing Weibo data with Python + UV.

    444 GitHub stars~1.8k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed

More from BlackBeltTechnology/pi-agent-dashboard

All 66 skills in this repo

  • Pi Dashboard

    BlackBeltTechnology/pi-agent-dashboard

    Monitor and control the pi-dashboard server. An agent skill from BlackBeltTechnology/pi-agent-dashboard.

    315 GitHub starsUsed in 1 repo~2.2k tokens
    Auto-check passed

  • CI Troubleshoot

    BlackBeltTechnology/pi-agent-dashboard

    Diagnose failed GitHub Actions runs for pi-agent-dashboard: the 11-file workflow taxonomy, affected-test selection, the release pipeline, known failure modes, and how to read gh run logs and…

    315 GitHub stars~3.5k tokensUpdated today
    Auto-check passed

  • Debug Dashboard

    BlackBeltTechnology/pi-agent-dashboard

    Diagnose problems in the running pi-agent-dashboard system: server.log, /api/health, bridge WebSocket connectivity, vitest triage, known-issue FAQ entries.

    315 GitHub stars~1.6k tokensUpdated today
    Auto-check passed

  • Implement

    BlackBeltTechnology/pi-agent-dashboard

    Disciplined implementation in pi-agent-dashboard: the rebuild matrix (extension→reload, server→restart, client→build+restart, openspec-apply→full rebuild) plus the project's code discipline rules.

    315 GitHub stars~2k tokensUpdated today
    Auto-check passed

  • Session To Guideline

    BlackBeltTechnology/pi-agent-dashboard

    Turn a pi session into a Markdown "how-we-did-it" collaboration guideline: reads the session's JSONL transcript and synthesizes a reusable playbook of which prompts worked, what had to be steered…

    315 GitHub stars~3.2k tokensUpdated today
    Auto-check passed

  • Bpmn Package Explorer

    BlackBeltTechnology/pi-agent-dashboard

    Generate, validate and view BPMN 2.0 process packages from a prose description, and render existing .bpmn / .dmn files.

    315 GitHub stars~1.7k tokensUpdated today
    Auto-check passed

Categories

Questions about Browser

What does Browser do?

Browser automation via the agent-browser CLI. An agent skill from BlackBeltTechnology/pi-agent-dashboard. Browser is an agent skill from BlackBeltTechnology/pi-agent-dashboard. Browser automation via the agent-browser CLI.

When should I use Browser?

Browser fits situations like: the user needs to drive websites; electron apps — navigating; data extraction; testing web apps.

How do I install Browser in Claude Code?

Run `npx skills add BlackBeltTechnology/pi-agent-dashboard --skill browser -a claude-code`. Or copy the skill folder (packages/extension/.pi/skills/browser in BlackBeltTechnology/pi-agent-dashboard) into .claude/skills/browser in your project. Claude Code loads it when a task matches its description.

How do I install Browser in Codex?

Run `npx skills add BlackBeltTechnology/pi-agent-dashboard --skill browser -a codex`. Or copy the skill folder (packages/extension/.pi/skills/browser in BlackBeltTechnology/pi-agent-dashboard) into .agents/skills/browser in your project. Codex loads it when a task matches its description.

Can I use Browser in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add BlackBeltTechnology/pi-agent-dashboard --skill browser -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/browser, .gemini/skills/browser, .github/skills/browser and .opencode/skills/browser in your project.

What does Browser need to run?

Going by SKILL.md and its folder, Browser needs a shell for the scripts in its folder and the command-line tools its instructions call (npm, curl and npx). Our summary lists: Node.js; A Bash shell. Its frontmatter pre-approves these tools: Bash(agent-browser:*), Bash(npx agent-browser:*), Bash(npx @panerelay/setup:*), Bash(curl:*).

Does Browser access the network?

SKILL.md contains no URLs. Its commands use npm, curl and npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Browser safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Browser use?

Browser is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Browser use?

About 2k tokens (SKILL.md is roughly 8.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 38k tokens, read only when the agent opens those files.

What are the alternatives to Browser?

Skills that share tags, products or a category with Browser: Zhihu Publish (bifrost-proxy/bifrost, 160 stars), Agent Browser (HHU3637kr/skills, 145 stars), Bitwarden Browser Aac Login (aiskillstore/marketplace, 430 stars) and Sap Browser Automation (secondsky/sap-skills, 462 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Browser?

BlackBeltTechnology (a GitHub organization) maintains it in BlackBeltTechnology/pi-agent-dashboard, which has 315 GitHub stars. The repository holds 66 skills in this directory. The repository was last updated on October 8, 2026.

Source: BlackBeltTechnology/pi-agent-dashboard on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.