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 agentmods add skills/ssube/conclave/browser-connectnpx skills add ssube/conclave --skill browser-connectgit clone --depth 1 https://github.com/ssube/conclaveWhat 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 | $0.00044 | $0.00514 |
| Opus 5 | $0.00022 | $0.00257 |
| Sonnet 5 | $0.00009 | $0.00103 |
| Haiku 4.5 | $0.00004 | $0.00051 |
Grade A, and why
browser-connect scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Verify Chromium is running: `curl http://127.0.0.1:9222/json/version` What it actually says
Browser Connect Skill
Connect to the local Chromium browser via CDP for automation. The browser runs
inside Conclave with CDP enabled on 127.0.0.1:9222.
Usage
Test the connection
python3 {baseDir}/browser_connect.py test
Take a screenshot
python3 {baseDir}/browser_connect.py screenshot [--output /path/to/screenshot.png]
List cookies
python3 {baseDir}/browser_connect.py cookies
Python API
from browser_connect import BrowserSession
with BrowserSession() as (browser, page):
page.goto("https://example.com")
page.screenshot(path="screenshot.png")
Environment Variables
CDP_HOST: CDP host (default:127.0.0.1)CDP_PORT: CDP port (default:9222)
Responsible Browsing
When using browser-connect to interact with websites, be a responsible internet citizen:
- Bookmark frequently visited pages — save and organize URLs you access often in your memory rather than navigating from scratch each time.
- Pace your requests — wait 2-3 seconds between clicking buttons and loading pages. Do not rapid-fire navigations.
- Avoid redundant loads — save screenshots or HTML content to analyze offline rather than loading the same page repeatedly.
- You have a semi-stealth browser with a real display (N.eko desktop), but you may need help with CAPTCHAs if they appear. Alert the user if you encounter one.
These guidelines apply to ALL skills that use browser-connect, as well. By following them, you help ensure a smoother experience for yourself and others while browsing the web.
Troubleshooting
Connection refused
- Verify Chromium is running:
curl http://127.0.0.1:9222/json/version - Check the N.eko desktop service is started
Page not loading
- The browser shares state with the N.eko desktop — if a user is active, pages may already be open
- Use
page.goto()to navigate to the desired URL
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 68 lines · 44 tokens per session scan A 76f8688d58b0
browser-connect is a skill published in the GitHub repository ssube/conclave (2 stars, last pushed 6mo ago), licensed MIT. It adds 44 tokens to every session and 514 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
playwright
Browser automation, web scraping, E2E testing, and visual regression with Playwright. Covers 30+ patterns: login flows, form testing, responsive design checks, broken link validation, API mocking, data extraction, PDF generation, accessibility audits (axe-core), performance budgets (Lighthouse), visual diffing…
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a…
tlamatini-daily-chat-test
Run the daily automated Tlamatini chat regression — drive a visible Chrome via Playwright, log into agentpage.html, ask up to 1000 curated safe questions one-by-one (Multi-Turn ON, ACPX/Ask-Execs/Exec-Report/Internet OFF), wait for and qualify each answer (heuristic + LLM judge on failures), then write a dated report…
webapp-testing
Toolkit for testing local web applications and browser workflows with MCP browser tools. Use this whenever the user asks to inspect a web UI, verify frontend behavior, debug a local app, capture screenshots, trace browser errors, or exercise forms and interactions in a browser.
browser-session-defense
Use this when the work is about security boundaries around Guardian-managed browser tools rather than generic web UI testing.
testsprite-onboard
Stand up a complete, runnable TestSprite test suite for the current repo at first use — create a project (with a target URL and auth), derive a coherent set of tests from the codebase, batch-create them, and smoke-run a few to a green verdict so the user immediately has something worth running. Use ONLY when a repo…