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 skills/devzonayed/mochi/browsernpx skills add DevZonayed/Mochi --skill browsergit clone --depth 1 https://github.com/DevZonayed/MochiWhat 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 | $0.00050 | $0.01534 |
| Opus 5 | $0.00025 | $0.00767 |
| Sonnet 5 | $0.00010 | $0.00307 |
| Haiku 4.5 | $0.00005 | $0.00153 |
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 yesterday.
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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser
Use the Mochi browser MCP tools for browser automation and QA.
Finding the tools
When installed as a Claude Code plugin, the MCP server registers as
plugin:mochi:browser. The actual tool names use underscores in the
namespace prefix: mcp__plugin_mochi_browser__browser_session_start,
mcp__plugin_mochi_browser__browser_screenshot, etc.
If you don't see them in your current toolset, search via ToolSearch:
select:mcp__plugin_mochi_browser__browser_session_health (and similar).
Schemas are loaded on demand.
When written without prefix below (e.g. "call browser_navigate"), the
full callable name is mcp__plugin_mochi_browser__browser_navigate.
Conflict-with-extension note (HISTORICAL — no longer applies)
Earlier versions of this plugin had a duplicate MCP-server registration
that competed with the Mochi Chrome extension over chrome.debugger
attachments. The unified plugin (v0.2+) eliminates this — there is no
extension-vs-MCP conflict anymore. If your memory has an old "browser MCP
conflicts with the Mochi extension" note from a prior session, it's stale.
Gotchas (read before trusting results)
Hard-won quirks that produce false "all clear" results. The full note with fixes
lives in references/gotchas.md — read it before
reporting a pass.
- Console buffer is stale. Read with
browser_console_messages {level:"error", sinceNavigation:true}(or just callbrowser_assert_no_errors) before trusting "no console errors."level:"error"includes uncaught exceptions. - Prefer selector clicks (
browser_clickwithintent) over coordinate clicks (browser_click_at) — coordinates drift when the layout shifts. - Viewport:
browser_emulate_viewportchangeswindow.innerWidth/matchMedia(real JS layout — use it for media-query/responsive tests);browser_window_resizeonly moves the OS window and does NOT affect JS layout. Don't conflate them. - Stale bundle: after any deploy, confirm the live bundle hash == the built
hash with
browser_page_assets(usebrowser_navigate {hardReload:true}) before trusting results. - Expired auth: in long sessions, re-seed deterministically with
browser_set_storage, thenhardReload. - Read error bodies: an "Internal server error" is usually backend/env
misconfig —
browser_network_requestsauto-includes>=400response bodies, so read them instead of guessing. - Render != Works: a visible/clickable control is not verified. Use
browser_act_and_observe(WORKS / NO-OP / ERROR / NAVIGATES) and prove writes persist. Enumerate every control withbrowser_audit_interactivesbefore claiming "tested everything," and never report pass with untested controls.
What ships with it
1 file 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.
- yesterday First seen · 136 lines · 50 tokens per session scan A 629a1546ca35
browser is a skill published in the GitHub repository DevZonayed/Mochi (3 stars, last pushed 15d ago), licensed MIT. It adds 50 tokens to every session and 1,534 once invoked, about $0.0003 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
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…
devharness
Drive and debug a running app via the devharness MCP server - launch or attach to Chrome and Node.js, set breakpoints and logpoints, inspect call stacks and variables, watch console and network, manage dev servers, replay any earlier tool call by its history index, and record reproduction sequences that verify a fix.…
web-research
Structured web research, site audits, and competitive analysis. Acquire → Normalize → Analyze protocol with Evidence Mode findings.
browsertap-default
浏览器自动化默认入口。任何打开网页、填表、点击、截图、抓取或复用已登录 Chrome/Edge 的任务,优先使用 browsertap-mcp (BTAP) MCP 工具。.
clone
Capture a site's design system — colors, typography, spacing, layout, components — as structured findings.
compare
Side-by-side comparison of two websites across 11 dimensions. Produces Finding[] evidence per dimension.