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 Nimm0ny/kimi-webbridge-mcp --skill kimi-webbridgegit clone --depth 1 https://github.com/Nimm0ny/kimi-webbridge-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/nimm0ny/kimi-webbridge-mcp/kimi-webbridge)<a href="https://agentmods.dev/skills/nimm0ny/kimi-webbridge-mcp/kimi-webbridge"><img src="https://agentmods.dev/badge/skills/nimm0ny/kimi-webbridge-mcp/kimi-webbridge/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/nimm0ny/kimi-webbridge-mcp/kimi-webbridge"><img src="https://agentmods.dev/badge/skills/nimm0ny/kimi-webbridge-mcp/kimi-webbridge.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 | $0.00215 | $0.02272 |
| Opus 5 | $0.00108 | $0.01136 |
| Sonnet 5 | $0.00043 | $0.00454 |
| Haiku 4.5 | $0.00021 | $0.00227 |
Grade A, and why
kimi-webbridge scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Prefer this over curl, ad-hoc shell, playwright scripts, or chrome-devtools when the task needs the How it starts
The opening of the file, as written. The whole thing — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kimi WebBridge
Skill = thin playbook. MCP = tools. Default profile is compact (Claude-in-Chrome sized). Prefer accuracy over calling many tools.
0. Agent 主动建链（硬前置 · 实测）
建链主体是 Agent，不是用户去点扩展。
用户侧扩展一般早已装好；MCP 挂上 ≠ 本任务已连通。Agent 必须在业务前 主动 调工具把链路拉起来。
链路分层
Agent wb_*
→ MCP (kimi-webbridge) ensureDaemon / command
→ 本机 daemon :10086 kimi-webbridge start（按需）
→ 浏览器扩展（通常已安装） extension_connected
→ 真实浏览器标签 / 登录态
| 层 | 谁拉起 | 标志 |
|---|---|---|
| MCP 工具面 | 宿主已配好 | 能调 wb_* |
| daemon | Agent 调 wb_status → MCP ensureDaemon（未跑则 kimi-webbridge start） |
running=true；daemon_started_now 可 true |
| 扩展 ↔ daemon | daemon 起来后扩展自动挂上（扩展已装好时） | extension_connected=true |
| 任务 session / 标签组 | Agent wb_navigate（group_title + 可选 session） |
可 wb_list_tabs |
| 站点登录 | 用户曾登录 / 本任务遇登录页再请用户 | 业务页非 login |
Agent 强制流程（每次要控浏览器时）
1) wb_status # 主动建链入口：会 ensureDaemon
2) 若 ready=true：
→ wb_navigate（首开传 group_title；业务 session 固定）
→ 后续 find/snapshot/click…
3) 若 ready=false 或 extension_connected=false：
→ 先 **Agent 侧重试**（勿立刻甩锅用户）：
- 再调 1～3 次 wb_status（daemon 刚 start 时扩展可能晚几秒连上）
- 仍 false：可 shell `kimi-webbridge status` / `kimi-webbridge start` 后复检 status
→ 仅多次失败才问用户：浏览器是否已打开（扩展是否在日常 profile 启用）
4) 禁止：未 wb_status 就假设可用；禁止默认话术「请你去点扩展连接」代替 Agent 建链
实测备忘（2026-08-11）
| 步骤 | 结果 |
|---|---|
| 仅 MCP 列出 / 未调工具 | 未视为已连接 |
wb_status |
MCP ensureDaemon；可 daemon_started_now=true；目标 ready=true |
ready 且 extension_connected |
扩展已装场景下，daemon 起后通常为 true |
wb_navigate + group_title + session |
打开业务页 / 建立标签组（例：OA 探针成功） |
| 扩展未装或浏览器未开 | status 持续 extension_connected=false → 再请用户处理 |
实现锚点（MCP）：ensureDaemon() 在 client.js；wb_status 在 index.js 内先 ensureDaemon 再返回 ready。
wb_status 字段
| 字段 | 含义 |
|---|---|
running |
daemon 是否在跑 |
daemon_started_now |
本次 wb_status 是否刚拉起 daemon（Agent 建链痕迹） |
extension_connected |
扩展是否已挂到 daemon |
ready |
running && extension_connected — 业务门槛 |
hint |
Ready 或未连时的提示 |
default_session |
默认 session 名 |
tool_profile / tool_count |
compact vs full |
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.
- 4d ago Changed · +10 lines ff0927f2bf1b
- 11d ago First seen · 140 lines · 215 tokens per session scan A 680228065871
kimi-webbridge is a skill published in the GitHub repository Nimm0ny/kimi-webbridge-mcp (0 stars, last pushed 5d ago), licensed MIT. It adds 215 tokens to every session and 2,272 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
interactive-login
How to complete browser/interactive logins (aws / gh / glab / gcloud). The platform backgrounds the login poller so it survives the human's browser round-trip — and when that does NOT work.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
azure-messaging-webpubsub-java
Build real-time web applications with Azure Web PubSub SDK for Java. Use when implementing WebSocket-based messaging, live updates, chat applications, or server-to-client push notifications.
google-safe-browsing
Prevent and fix Google Safe Browsing "Dangerous site" flags. Use when launching a public web app, buying/picking a domain, building a login or signup page, or when any site shows a red "Dangerous site" / "Deceptive site" warning in Chrome, Brave, Safari, Firefox, or Edge. Triggers on "dangerous site", "deceptive…