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 wavyrai/tmux-ide --skill herdr-gpui-browsergit clone --depth 1 https://github.com/wavyrai/tmux-ideWrote 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/wavyrai/tmux-ide/herdr-gpui-browser)<a href="https://agentmods.dev/skills/wavyrai/tmux-ide/herdr-gpui-browser"><img src="https://agentmods.dev/badge/skills/wavyrai/tmux-ide/herdr-gpui-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/wavyrai/tmux-ide/herdr-gpui-browser"><img src="https://agentmods.dev/badge/skills/wavyrai/tmux-ide/herdr-gpui-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 | $0.00109 | $0.01336 |
| Opus 5.5 | $0.00044 | $0.00534 |
| Sonnet 5.5 | $0.00022 | $0.00267 |
| Haiku 4.5 | $0.00011 | $0.00134 |
Grade A, and why
herdr-gpui-browser 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 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.
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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser tabs in Herdr GPUI
Inside a Herdr pane (HERDR_ENV=1), "open a browser tab" means a tab in Herdr
GPUI, beside the user's terminals: use the commands below, not open, a
system browser, or another tool's browser skill.
Herdr GPUI, the native desktop client for Herdr, can show web pages in browser tabs that sit beside the terminal tabs of a workspace. The user can annotate a page, pinning notes to elements or selected text, and send the notes back to you. Use it when the user should see a page rather than read a URL, for example:
- a page, mockup, or report you generated as an HTML file, for review
- a local dev server or preview you just started (
http://localhost:3000) - a pull request, issue, CI run, or deployment you created
- documentation or a design you are referring to
Open a page
herdr-gpui browser open ./mockup.html
herdr-gpui browser open http://localhost:3000
- The tab opens in your own workspace (
HERDR_WORKSPACE_ID, which Herdr sets in every pane) and the window switches to it. The terminal keeps running underneath; the user returns to it by clicking its tab. - A local file is served with the other files in its folder, so relative links to CSS, scripts, and images work. Keep the page and its assets in a dedicated folder: never your home directory, and nothing in a hidden folder. Files whose names start with a dot are never served.
- Opening the same page again from the same pane reuses and reloads its tab.
After editing a file, run
herdr-gpui browser reloadinstead. - Add
--no-focusto add the tab without switching to it, for example when the user is typing in your pane. - Bare hosts work:
localhost:3000becomeshttp://localhost:3000/, andexample.com/docsbecomeshttps://example.com/docs. Onlyhttp,https, and local files open.
Tell the user what you opened and why, in one line. Open a page when it helps the user, not after every step.
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 · 103 lines · 109 tokens per session scan A e90df6d27974
herdr-gpui-browser is a skill published in the GitHub repository wavyrai/tmux-ide (550 stars, last pushed today), licensed MIT. It adds 109 tokens to every session and 1,336 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-15.
Other skills, from other repositories
computer-use
Desktop control that picks the right interface per step — app scripting (AppleScript/JXA), accessibility-first observation and actions, pixel fallback, screenshots, zoom, screen recording, and switching between registered computers. Qualified on macOS; Windows, Linux and HarmonyOS backends are experimental.
webapp-testing
Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
read-file
Extract the actual text/content from a file of ANY format — PDF, Word, PowerPoint, Excel, images, scans, archives. The canonical "how do I read this file" reference, reused by process-file, Pulse ingestion, and browser downloads.
basic-browser-setup
Set up browser QA in AgentWorks, save verified locators and application knowledge, configure a repeatable smoke test, and establish reporting. Use when onboarding an application or repairing its QA foundation.
browser-performance-validation
Build repeatable AgentWorks browser performance validation with customer-defined budgets and comparable samples. Use for page, journey, Web Vital, resource, or timing regressions.
browser-use
Control the user's Chrome through the existing persistent Node REPL and the Qwen Browser SDK.