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 steerlabs/opensteer --skill browsergit clone --depth 1 https://github.com/steerlabs/opensteerWrote 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/steerlabs/opensteer/browser)<a href="https://agentmods.dev/skills/steerlabs/opensteer/browser"><img src="https://agentmods.dev/badge/skills/steerlabs/opensteer/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/steerlabs/opensteer/browser"><img src="https://agentmods.dev/badge/skills/steerlabs/opensteer/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 | $0.00021 | $0.00416 |
| Opus 5.5 | $0.00008 | $0.00166 |
| Sonnet 5.5 | $0.00004 | $0.00083 |
| Haiku 4.5 | $0.00002 | $0.00042 |
Grade A, and why
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 today.
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.
What it actually says
Opensteer Browser
Use Browser Sessions for website interaction and saved pages. For account/API work, use opensteer:integrations unless the person chose the website or no applicable integration exists. The Host owns sessions, profiles and grants; Browser Harness operates pages.
Choose the identity yourself, then observe the exact page: Guest for signed-out work, a listed profile that fits for anything done as the person, and browser_profile_request only when none fits. Only the person creates or grants profiles. Profile names do not prove a website account. Never substitute Guest when a profile fails to open, or sign in there for work the person expects to keep.
With full access, browser_profile_write lists profiles and grants, renames or deletes profiles, clears history or remembered sites, and revokes grants directly. Request sign-in with sign_in; never manage sign-in sessions or retrieve saved credentials yourself. SMS/email codes for a sign-in the person asked for may come from an already-granted integration; follow the sign-in reference.
On a voice call, show pages worth following as you start reading or working in them and when you move to another; a shown page stays visible, so skip repeat shows and quick lookups.
Use the supplied Browser Harness for page operations. Preserve the person's pages and arrangement, and inspect the outcome before repeating input that may have happened.
- Read identity before choosing, opening or requesting a profile or session.
- Read cards before using a saved payment card at checkout.
- Read sign-in when a website needs authentication.
- Read tabs and Show when finding, organizing, using an included page or showing one to the person.
- Read page API before writing Harness code or handling page input, uploads, sizing or dialogs.
What ships with it
6 files 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.
- today Changed · +3 lines a4f4429b8091
- yesterday Changed · +2 lines · -5 tokens per session c6ac2fc66b84
- 2d ago First seen · 18 lines · 26 tokens per session scan A a2931bd0251e
browser is a skill published in the GitHub repository steerlabs/opensteer (204 stars, last pushed today), licensed MIT. It adds 21 tokens to every session and 416 once invoked, about $0.0001 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 2026-11-04.
browser compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| browsersteerlabs/opensteer | Skill | 204 | today | MIT | A | 21 | Codex |
| kortix-computerkortix-ai/suna | Skill | 20k | 3d ago | none | A | 271 | Claude Code, Codex |
| weave-browserTyox-all/Weave_Protocol | Skill | 1 | 2d ago | Apache-2.0 | A | 144 | Claude Code, Codex |
| webapp-testingThreat-Vector-Security/guardian-agent | Skill | 13 | 1mo ago | Apache-2.0 | A | 55 | Claude Code, Codex |
| browser-session-defenseThreat-Vector-Security/guardian-agent | Skill | 13 | 1mo ago | Apache-2.0 | A | 0 | Claude Code, Codex |
| n8n-clin8n-io/n8n | Skill | 210k | yesterday | none | A | 55 | Claude Code |
| suggest-automationsn8n-io/n8n | Skill | 210k | yesterday | none | A | 87 | Claude Code, Codex |