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 skills add MrTrenchTrucker/contak-browser-skills --skill kasm-browsergit clone --depth 1 https://github.com/MrTrenchTrucker/contak-browser-skillsWrote 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.
[](https://agentmods.dev/skills/mrtrenchtrucker/contak-browser-skills/kasm-browser)<a href="https://agentmods.dev/skills/mrtrenchtrucker/contak-browser-skills/kasm-browser"><img src="https://agentmods.dev/badge/skills/mrtrenchtrucker/contak-browser-skills/kasm-browser/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mrtrenchtrucker/contak-browser-skills/kasm-browser"><img src="https://agentmods.dev/badge/skills/mrtrenchtrucker/contak-browser-skills/kasm-browser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.00634 |
| Opus 5 | $0.00010 | $0.00317 |
| Sonnet 5 | $0.00004 | $0.00127 |
| Haiku 4.5 | $0.00002 | $0.00063 |
Grade A, and why
kasm-browser 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 10d 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.
1. Confirm CDP reachable: `docker exec <container> curl -s http://localhost:9222/json/version`. How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
KASM Browser via CDP
When invoked, drive a persistent KASM Chrome browser using Chrome DevTools Protocol. Use this when a flow needs login state, cookies that persist, captcha/2FA, or React-style synthetic events.
Inputs
action-- one of:navigate,click,type,screenshot,close_tab,evalurl-- target URL (fornavigate)selector-- CSS selector (forclick,type)text-- text to type (fortype)tab_id-- existing tab to target (optional; otherwise pick or create)output_path-- forscreenshotaction, where to save the PNG
Output
screenshotaction: file path to PNG.- Other actions: status + DOM snapshot or evaluated value.
Steps
- Confirm CDP reachable:
docker exec <container> curl -s http://localhost:9222/json/version. - List tabs:
docker exec <container> curl -s http://localhost:9222/json. - Pick or create the target tab; capture its
webSocketDebuggerUrl. - Open a WebSocket to that URL from inside the container.
- Send the CDP commands for the action, applying the Human-Speed Rule between every meaningful step.
- Capture results (DOM, screenshot, console output).
- Close the WebSocket; tidy up any tabs your flow created.
The Human-Speed Rule (mandatory)
Slow IS fast.
- Minimum 1.5-3 seconds between meaningful actions.
- Typing: one key at a time, 80-150 ms between keystrokes with jitter. Do NOT blast a full string via
input.value = "..."on sites that profile typing cadence. - Wait for
Page.loadEventFiredbefore queuing a second navigation. - Self-imposed rate limit: ~20 CDP calls per 10 seconds.
Bot-detection systems flag burst patterns even when intent is benign. Human-speed pacing is free immunity. Apply regardless of how impatient the user is.
Example: navigate + read title
const sock = new ws(tab.webSocketDebuggerUrl);
// ... open WS, set up send() helper ...
await send("Page.navigate", { url: "https://example.com" });
await new Promise(r => setTimeout(r, 3000)); // human-speed wait
const title = (await send("Runtime.evaluate", { expression: "document.title" })).result.result.value;
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.
- 10d ago First seen · 54 lines · 19 tokens per session scan A c4d2a0846983
kasm-browser is a skill published in the GitHub repository MrTrenchTrucker/contak-browser-skills (5 stars, last pushed 4mo ago), licensed MIT. It adds 19 tokens to every session and 634 once invoked, about $0.0001 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
pinchtab
Use this skill when a task needs browser automation through PinchTab: open a website, inspect interactive elements, click through flows, fill out forms, scrape page text, reuse a dedicated automation profile with user approval, export screenshots or PDFs, manage multiple browser instances, or fall back to the HTTP API…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
Agent Browser Automation
Fast Rust-based headless browser automation CLI with Node.js fallback for AI agents, featuring navigation, clicking, typing, snapshots, and structured commands optimized for agent workflows.
owb
Open Web Bridge (OWB) — drive the user's own real browser with the owb command. Read pages behind their existing logins, gather and cross-check information, fill forms, walk multi-step flows, debug their site, audit responsive/accessibility behavior, and capture or reverse-engineer network traffic. Use this whenever…
chrome-agent
Local browser automation with structured, verified outcomes. Use for web navigation, scraping and extraction, form interaction, screenshots and downloads, network or console checks, responsive testing, or page-scoped device emulation.
scrape-structured-data
Get the repeating records off a web page (product grids, search results, job listings, news feeds, tables) as JSON, without writing CSS selectors and without spending a model call to read the HTML. Works on sites with no API, including ones behind a login or bot protection. Runs locally, one binary, no API key. Use…