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 LogicLabsAI/ultramemory-mcp --skill playwright-human-modegit clone --depth 1 https://github.com/LogicLabsAI/ultramemory-mcpWrote 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/logiclabsai/ultramemory-mcp/playwright-human-mode)<a href="https://agentmods.dev/skills/logiclabsai/ultramemory-mcp/playwright-human-mode"><img src="https://agentmods.dev/badge/skills/logiclabsai/ultramemory-mcp/playwright-human-mode/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/logiclabsai/ultramemory-mcp/playwright-human-mode"><img src="https://agentmods.dev/badge/skills/logiclabsai/ultramemory-mcp/playwright-human-mode.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 | $0.00157 | $0.03500 |
| Opus 5 | $0.00078 | $0.01750 |
| Sonnet 5 | $0.00031 | $0.00700 |
| Haiku 4.5 | $0.00016 | $0.00350 |
Grade A, and why
playwright-human-mode scanned grade A with 0 findings 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 11d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser Control — Human Vision Mode (Supervised Co-Working)
UltraMemory Agent Kit — our Playwright Human Vision Control wrapper over the standard Playwright MCP. The kit installs the standard Playwright MCP (on your consent) and this skill rides on top, adding the speed mandate + supervised co-working discipline below.
⚡ SPEED MANDATE — read before any browser action (HARD RULE, do not drift)
This skill is only as fast as you drive it. The default failure mode is self-inflicted slowness: one action per round-trip and a screenshot + narration after every click. DO NOT. Near-real-time is the bar.
Non-negotiable defaults, every browser session:
- BATCH. Put independent actions in ONE message, and chain a known UI sequence without stopping between each — e.g. copy password → focus field → paste → submit is ONE message, not four. One round-trip, not five.
- Snapshot-first, sparingly. Act off
browser_snapshot(the a11y text tree). Do NOT re-snapshot after every action — refs stay valid until the DOM changes; only re-snapshot when the page actually changed. Screenshot only at checkpoints / when pixels matter. - Reuse the session.
browser_navigatestraight to the target and CHECK if you're already authenticated before ever touching a login form. - Narrate per phase, not per click — "Logging in…", "At consent" — then move.
- Gates only at decisions / irreversible / money / login-2FA handoff. Everywhere else, batch and go. Supervision ≠ a click-by-click play-by-play.
browser_evaluatefor compound / multi-step page work. For forms and dense admin UIs, ONEbrowser_evaluatecan find + fill + select + click + read in a single call — far fewer round-trips than snapshot→click→snapshot, and it avoids generating huge page snapshots. Reach for it whenever a step is multi-part or a full snapshot would be a wall of text. (Keep secrets out of the return: act on the value in-page and return only a confirmation, e.g. prefix + length.)
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.
- 11d ago First seen · 225 lines · 157 tokens per session scan A 4850a8adbc8d
playwright-human-mode is a skill published in the GitHub repository LogicLabsAI/ultramemory-mcp (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 157 tokens to every session and 3,500 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
e2e-testing-patterns
Master end-to-end testing with Playwright and Cypress to build reliable test suites that catch bugs, improve confidence, and enable fast deployment. Use when implementing E2E tests, debugging flaky tests, or establishing testing standards.
agent-browser
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
ultimate-browsing
Reaches web pages a plain fetch cannot: renders JS, drives clicks and forms, captures screenshots, holds a login, and gets past WAF blocks through platform-native readers and stealth Chrome. Use for any page work beyond retrieving static text.
dev-browser
Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website"…
dashboard
Open OwnMem Console, the local dashboard for this repository's memory. Use when the user asks to open the dashboard, see memory metrics, check adoption or recall quality, or set up the optional embedding lane. Requires a repository initialized with the dashboard layer.
recall
Recall this repository's OwnMem local memory before changing code, and keep it healthy. Use when a repository contains .ownmem/, when past debugging lessons could apply ("have we hit this before", "why is it done this way"), or when the user mentions ownmem, project memory, or recalling across sessions.