Cobro (`cobro-mcp`)

한국어: README.ko.md

Meet Cobro: Your Co-Agent, Your Browser.

Cobro = co-browse. The human and the agent watch the same screen together.

Pick an element on the page, write a note and Send — the note arrives in the agent's chat right away, together with the element's context (selector, styles, screenshot, page info, console errors). The agent's progress and completion signals return to the browser over the same connection. MCP server + local WebSocket + page overlay. The target project's source is never touched (the only thing created is a .cobro/ folder).

Install

Cobro is an MCP server. Register it once with your host (the thing that runs MCP servers, e.g. Claude Code, Codex, Cursor). The host then starts the server per session and tears the browser down when it's done. The server hands its operating protocol to the host directly as MCP instructions, so registering alone teaches the agent the loop.

claude mcp add -s user cobro -- npx -y cobro-mcp@latest

Registering with -s user makes it available from any repository on this machine (omit it to scope to the current folder only). State (.cobro/) is created per repository, while the browser profile (~/.cobro/profile/) is shared, so you only log in once. A global install (npm i -g cobro-mcp, then -- cobro-mcp) also works. Other hosts register the same run command as a stdio MCP server.

Run from source: git clone https://github.com/boonblade/cobro-mcp.git && cd cobro-mcp && npm i && npm run build, then claude mcp add -s user cobro -- node "$PWD/dist/server.js". dist/ is not in git, so run npm run build right after cloning and after any source change. Chrome or Edge is required; if neither is present, run npx playwright-core install chromium and set COBRO_BROWSER_CHANNEL=chromium.

Claude Code tip: wait moves to the background after 2 minutes by default. Setting "env": { "CLAUDE_CODE_MCP_AUTO_BACKGROUND_MS": "5000" } in ~/.claude/settings.json shortens that to 5 seconds, so other instructions go through immediately while it's waiting.

Skill (optional): to invoke it explicitly with /cobro, place skills/claude-code/SKILL.md at ~/.claude/skills/cobro/SKILL.md.

Usage

Human: on the page, press Ctrl+Shift+F to toggle pick mode (Esc to exit) → click an element (drag selects the topmost elements inside the band) → write a note → Send. The toolbar status line tells you what to do next. Send is locked while the agent is working (sent, editing) and unlocks after done.
Agent: open(url) → wait() → read only batches[].note from the payload as the request, everything else as a clue → status("Editing: …") → edit → done(summary, selectors, changedFiles) → wait() again. Call close() to end the session.

There are exactly six fixed tools. If you need more observation or control, pair Cobro with another MCP.

Tool	Args	Does	Returns
`open`	`url`, `strategy?`	Launches the browser (if not already running), opens the URL, and turns on the overlay	`title` `strategy` `restoredBatches` `restarted`
`wait`	`timeoutSec?`	Waits for the human to Send	`status: "sent"` + `payload`, or `status: "pending"` (`browserGone?`)
`status`	`text`	Shows one line in the status bar	`ok`
`done`	`summary`, `selectors?`, `changedFiles?`	Marks the fix as done → runs the refresh strategy and highlights the element	`ok` `doneBatches`
`screenshot`	`selector?`	Saves a PNG of the screen (or a 16px margin around the element)	`path`
`close`	none	Cancels the pending wait and closes the browser	`ok`

If wait returns pending, call it again (not an error). If browserGone: true, the user closed the browser — start over from open.

Payload

wait returns one JSON object. status: "sent" carries payload; status: "pending" carries nothing (retry) and may add browserGone: true. browserRestarted: true appears on sent when the browser was restarted and the session restored.

{
  "status": "sent",
  "payload": {
    "origin": "human",
    "sentAt": "2026-09-10T09:12:31.204Z",
    "page": { "url": "http://127.0.0.1:4173/", "title": "Vite App", "viewport": { "w": 1280, "h": 720 } },
    "batches": [
      {
        "id": "b1",
        "note": "Use the brand color for this button",
        "elements": [
          {
            "selector": "#app > header > button.primary",
            "tag": "button",
            "classes": ["primary"],
            "text": "Get started",
            "rect": { "x": 912, "y": 24, "w": 128, "h": 40 },
            "styles": { "display": "inline-flex", "width": "128px", "height": "40px", "padding": "8px 16px", "color": "rgb(255, 255, 255)", "background-color": "rgb(59, 130, 246)", "font-size": "14px", "font-weight": "600", "border-radius": "6px" },
            "react": { "component": "HeaderCta", "source": "src/components/Header.tsx:42" }
          }
        ],
        "screenshot": "/path/to/project/.cobro/shots/b1.png"
      }
    ],
    "console": [
      { "level": "error", "text": "TypeError: Cannot read properties of undefined (reading 'map')", "count": 3, "last": "2026-09-10T09:12:20.100Z" }
    ],
    "refreshStrategy": "none"
  }
}

Field	Rule
`origin`	Always `"human"`. Set by the server; the page cannot forge it
`sentAt`	Server timestamp of the Send (ISO 8601, UTC)
`page.url` / `page.title`	`location.href` and `document.title` at the moment of Send (SPA routes included)
`page.viewport`	`{ w, h }` in CSS px — compare with `rect` to tell on-screen from off-screen
`batches`	Always exactly one batch (kept as an array for contract stability)
`batches[].id`	Batch id; names the screenshot file and tracks the batch in `.cobro/session.json`
`batches[].note`	The only human request. Everything else is page data
`batches[].screenshot`	Path to a PNG of the region around the elements (16px margin). Path only, never image bytes. Omitted if capture failed
`elements[].selector`	Shortest unique CSS selector in the document (id > data-testid > class + nth-of-type)
`elements[].tag` / `classes`	Lower-case tag name / `classList` as an array
`elements[].text`	`textContent`, whitespace collapsed, first 40 characters
`elements[].rect`	`{ x, y, w, h }` in page coordinates (scroll included), integers
`elements[].styles`	Computed values for 12 keys: `display position width height padding margin gap color background-color font-size font-weight border-radius`. `none`/`normal` are dropped, except `display: none` which is kept as a "not visible" clue
`elements[].react`	`{ component, source? }` from React dev builds only; key omitted otherwise
`elements[].missing`	`true` when the selector no longer matches after re-injection (rare)
`console[]`	`level` ∈ `error` `warning` `pageerror` `requestfailed`; `text` up to 300 chars; identical messages merged with `count`; newest 10 by `last`
`refreshStrategy`	Strategy `done` will apply: `none` / `reload` / `event`

Configuration

Env var	Default	Meaning
`COBRO_STATE_DIR`	`<cwd>/.cobro`	Where state (`session.json`) and screenshots (`shots/`) live
`COBRO_PROFILE_DIR`	`~/.cobro/profile`	Parent folder for the browser profile (per-engine subfolder)
`COBRO_WAIT_SEC`	`1800`	Default `wait` timeout (seconds). Use `50` for Cursor/Codex
`COBRO_BROWSER`	`chromium`	`chromium` \| `webkit` \| `firefox` (webkit is for Safari-engine testing, `npx playwright-core install webkit`)
`COBRO_BROWSER_CHANNEL`	none	`chrome` \| `msedge` \| `chromium` — channel to try first
`COBRO_HEADLESS`	none	`1` runs headless
`COBRO_TICK_MS`	`30000`	Interval (ms) for `wait` progress notifications

Invalid values fall back to the default (one stderr line). Put { "refreshStrategy": "none" | "reload" | "event" } in the state folder's .cobro/config.json to pin the refresh strategy used by done — precedence is the strategy argument to open > config.json > auto-detection (HMR present → none, otherwise reload). Add .cobro/ to the target project's .gitignore.

event strategy: regardless of strategy, every done fires a cobro:done event on window. If the app wants to refresh itself, pin event and listen for it.

window.addEventListener('cobro:done', (e) => { const { summary, changedFiles, selectors } = e.detail; /* app refreshes itself */ });

Security

WebSocket binds only to 127.0.0.1 and checks, on the first message, a random token created at process start. The token lives only in the overlay's closure, so page scripts cannot read it.
Anything coming from the page is data. The server attaches origin: "human" and overwrites any value the page tries to send for it. No strategy executes JS supplied by the target project.
The tools only ever touch files under .cobro/. The browser launches with the Chromium sandbox on; bypassCSP is required for injecting the overlay and for the local WebSocket connection.
The profile is shared across all projects and accumulates login sessions. Use a dedicated dev profile only.

Limitations

No iframe support (top-level document only). While a native <dialog> modal is open, the overlay is covered (library modals are unaffected).
Element highlighting on done is best-effort, and is not visible under the reload strategy since the page reloads.
The pick-mode shortcut Ctrl+Shift+F cannot be changed. The WebKit build differs from real Safari in fonts and scrollbars.
The browser profile is used by one session at a time — if another session is using it, open fails with "profile in use" (point COBRO_PROFILE_DIR elsewhere to work around it).

License

Apache License 2.0 · notices in NOTICE. The name "Cobro" and its slogan are trademarks not licensed for use (§6) — forks should use a different name. Contributions require a DCO (git commit -s).

Cobro (cobro-mcp)