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 agents/undertone0809/rudder/built-in-browsergit clone --depth 1 https://github.com/Undertone0809/rudderWrote 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/agents/undertone0809/rudder/built-in-browser)<a href="https://agentmods.dev/agents/undertone0809/rudder/built-in-browser"><img src="https://agentmods.dev/badge/agents/undertone0809/rudder/built-in-browser.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.07557 |
| Opus 5 | $0.00000 | $0.03778 |
| Sonnet 5 | $0.00000 | $0.01511 |
| Haiku 4.5 | $0.00000 | $0.00756 |
Grade A, and why
built-in-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 4d 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 — 550 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rudder Built-in Browser
AGENT.BROWSER.001
Contract Summary
Rudder Desktop provides an instance-scoped Built-in Browser for operators and
Rudder-managed agents. It is enabled by default, opens ordinary Rudder web
links in a Browser tab in the global Side Panel by default, can transfer an
exact operator tab into the Messenger Main Workbench, and stores one
persistent website profile per operating-system user and canonical Rudder
instance. Organizations intentionally share that website identity, while Agent
Browser tabs and control leases remain isolated by organization, agent, run,
and tab. Explicit operator address-bar input may also bootstrap a narrowly
defined canonical local absolute file:/// target; renderer links, page-driven
navigation, and Agent Browser control remain HTTP(S)-only.
V1 is a Desktop local_trusted capability. It includes local macOS Chromium
cookie import, a conditional read-only Browser skill, and eight high-level
rudder_browser_* tools. It does not provide password import, arbitrary page
script execution, raw cookie access, or remote control of a local Desktop.
Intent / User Job
- Operators can inspect and use web content without losing their current Rudder work surface, choose the system browser when preferred, import an existing signed-in session, and reset the shared Browser profile.
- Agents can perform bounded browser navigation and interaction through a typed, runtime-owned control surface instead of receiving raw credentials, unrestricted CDP, or model-supplied browser identity.
Why / Design Reasoning
One profile per local instance makes Browser sessions useful across the whole Rudder workspace and avoids a separate login for every organization. That convenience is an explicit trust decision: settings and import confirmation must disclose cross-organization session sharing. Control remains run-scoped so shared cookies do not imply shared tab ownership.
The Browser skill and tools are one capability resolved from the live instance
setting. Rudder projects them at run time rather than writing permanent skill
assignments to every organization. High-level tools reuse the first-party
rudder-tools transport so authentication, runtime identity, and
auditing remain inside existing boundaries.
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.
- 4d ago First seen · 550 lines · 0 tokens per session scan A f8657b64ba82
built-in-browser is an agent published in the GitHub repository Undertone0809/rudder (288 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 7,557 tokens. 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 2026-08-30.
Other agents, from other repositories
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…
praman-sap-planner-cli
SAP UI5 test planner via Playwright CLI. Token-efficient alternative to MCP planner. Generates test plan + gold-standard spec using CLI commands.
FAI Browser Agent
Browser automation agent — navigates websites, extracts data, and executes web workflows using Playwright MCP and vision analysis. Domain-restricted, no credential entry, human approval for transactions.
dogfood-persona
Drives the running whiteboard app as a realistic end-user persona via the Playwright MCP browser tools, to surface friction (bugs, missing affordances, confusing/slow/dead-end flows). Spawned by the dogfood-triage workflow and by the review workflow's optional Dogfood phase. Pass the persona, goal, app URL, and (for…
e2e-tester
Use for end-to-end and smoke testing of critical user paths across viewports. Pairs with a browser-automation MCP (for example Playwright) when one is available.
judge-x-claude-code
Independent Gauntlet judge for Claude Code, invoked by the harness through the evaluator adapter (dispatch.ts --judge-x). Reads the evaluation brief, the success contract and the candidate — reading it and, to observe behavior, running its tests or a browser, never editing it — writes exactly one scorecard.json into…