* fix(pairing): reject reserved clientId 'root' at all token writers 'root' is the sentinel checkScope/checkDomain/checkRate and the server command gate use for the omnipotent caller, so a scoped token carrying it bypasses every enforcement path. Add ReservedClientIdError + a shared assertValidClientId; createToken/createSetupKey throw, restoreRegistry skips-and-logs (a corrupt state file must not brick boot). /pair and /token surface it as a named 400, and the CLI fast-fails --client root. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(pairing): release tab ownership on revoke tabOwnership cleared only on tab close, so after DELETE /token a same-name re-pair inherited the revoked agent's authenticated tabs (own-only access keys on owner === clientId). Add BrowserManager.releaseClientTabs and run it unconditionally in DELETE /token (ownership outlives the token, so an expired-token client can still own tabs); 404 only when both nothing was revoked and nothing released. Response now carries tabs_released. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * v1.68.3.0 fix(pairing): re-pair to narrow revokes the old grant on the spot POST /pair minted a new setup key but never touched the agent's live session, so re-pairing --client X --restrict read while X was connected (or whose 5-min key expired unexchanged) left the original full-access session, eval included, alive up to 24h. A reducing re-pair (fewer scopes, tighter domains, lower rate, stricter tab policy) now revokes the live session and releases its tabs before minting the new key (grantReducesAccess + revokeClientFully; superseded in the response). Non-reducing re-pairs keep the session and only drop stale PENDING setup keys, so a broaden/refresh never strands a working agent and a narrowing re-pair issued before the agent connects can't leave the old broad key exchangeable. Revoke happens before mint (revokeToken deletes all of a client's tokens). CLI prints a version-skew-safe supersede notice and warns when a re-pair-shaped call omits --client. Docs + CHANGELOG + VERSION. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(pairing): harden re-pair per adversarial review Adversarial review of the diff found four issues, now fixed: - Validate the requested grant BEFORE the supersede revoke: a reducing re-pair with a bad scope/rate no longer destroys the live session and then fails to mint a replacement (assertValidTokenOptions runs up front). - A re-pair with no live session releases tabs orphaned by an expired incarnation, closing the tab-inheritance gap /pair had (DELETE /token already released unconditionally). - Test the DELETE /token revoked=0/tabs>0 path and the /pair orphaned-tab release at the handler level (HTTP e2e can't, headless owns no tabs). - Test the CLI --client root fast-fail; fix its null-guard (parseFlag returns null when --client is absent). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Garry Tan <garry@ycombinator.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
12 KiB
Remote Browser Access — How to Pair With a GStack Browser
A GStack Browser server can be shared with any AI agent that can make HTTP requests. The agent gets scoped access to a real Chromium browser: navigate pages, read content, click elements, fill forms, take screenshots. Each agent gets its own tab.
This document is the reference for remote agents. The quick-start instructions are
generated by $B pair-agent with the actual credentials baked in.
Architecture
Your Machine Remote Agent
───────────── ────────────
GStack Browser Server Any AI agent
├── Chromium (Playwright) (OpenClaw, Hermes, Codex, etc.)
├── Local listener 127.0.0.1:LOCAL │
│ (bootstrap, CLI, sidebar, cookies) │
├── Tunnel listener 127.0.0.1:TUNNEL ◄───────┤
│ (pair-agent only: /connect and │
│ /command — locked allowlist) │
├── ngrok tunnel (forwards tunnel port only) │
│ https://xxx.ngrok.dev ─────────────────┘
└── Token Registry
├── Root token (local listener only)
├── Setup keys (5 min, one-time)
├── Session tokens (24h, scoped)
└── SSE session cookies (30 min, stream-scope)
Dual-listener architecture (v1.6.0.0)
The daemon binds two HTTP sockets. The local listener serves the full command surface to 127.0.0.1 only and is never forwarded. The tunnel listener is bound lazily on /tunnel/start (and torn down on /tunnel/stop) with a locked path allowlist. ngrok forwards only the tunnel port.
A caller who stumbles onto your ngrok URL cannot reach /health, /cookie-picker, /inspector/*, or /welcome — those paths don't exist on that TCP socket. Root tokens sent over the tunnel get 403. The tunnel listener accepts only /connect and /command (with a scoped token + the 26-command browser-driving allowlist).
See ARCHITECTURE.md for the full endpoint table.
Connection Flow
- User runs
$B pair-agent(or/pair-agentin Claude Code) - Server creates a one-time setup key (expires in 5 minutes)
- User copies the instruction block into the other agent's chat
- Remote agent runs
POST /connectwith the setup key - Server returns a scoped session token (24h default)
- Remote agent creates its own tab via
POST /commandwithnewtab - Remote agent browses using
POST /commandwith its session token + tabId
API Reference
Authentication
All command endpoints require a Bearer token:
Authorization: Bearer gsk_sess_...
/connect is unauthenticated (rate-limited) — it's how a remote agent exchanges a setup key for a scoped session token. /health is unauthenticated on the local listener (liveness/status only — never a token) but does NOT exist on the tunnel listener (404). Extension token bootstrap is POST /extension-token on the local listener, gated by the pinned chrome-extension:// Origin; it is not on the tunnel surface either.
SSE endpoints (/activity/stream, /inspector/events) accept either a Bearer token or the HttpOnly gstack_sse cookie (minted via POST /sse-session, 30-minute TTL, stream-scope only — cannot be used against /command). As of v1.6.0.0 the ?token=<ROOT> query-string auth is no longer accepted.
Endpoints
POST /connect
Exchange a setup key for a session token. No auth required. Rate-limited to 300/minute (flood defense — setup keys are 24 random bytes, unbruteforceable).
Request: {"setup_key": "gsk_setup_..."}
Response: {"token": "gsk_sess_...", "expires": "ISO8601", "scopes": ["read","write","admin","meta"], "agent": "agent-name"}
POST /command
Send a browser command. Requires Bearer auth.
Request: {"command": "goto", "args": ["https://example.com"], "tabId": 1}
Response: (plain text result of the command)
GET /health
Server status. No auth required. Returns status, tabs, mode, uptime.
Never carries a token — extension token bootstrap is POST /extension-token
(local listener only, validates the pinned chrome-extension:// Origin and a
loopback Host; 403 otherwise). Not reachable over the tunnel (404).
Commands
Navigation
| Command | Args | Description |
|---|---|---|
goto |
["URL"] |
Navigate to a URL |
back |
[] |
Go back |
forward |
[] |
Go forward |
reload |
[] |
Reload page |
Reading Content
| Command | Args | Description |
|---|---|---|
snapshot |
["-i"] |
Interactive snapshot with @ref labels (most useful) |
text |
[] |
Full page text |
html |
["selector?"] |
HTML of element or full page |
links |
[] |
All links on page |
screenshot |
["/tmp/s.png"] |
Take a screenshot |
url |
[] |
Current URL |
Interaction
| Command | Args | Description |
|---|---|---|
click |
["@e3"] |
Click an element (use @ref from snapshot) |
fill |
["@e5", "text"] |
Fill a form field |
select |
["@e7", "option"] |
Select dropdown value |
type |
["text"] |
Type text (keyboard) |
press |
["Enter"] |
Press a key |
scroll |
["down"] |
Scroll the page |
Tabs
| Command | Args | Description |
|---|---|---|
newtab |
["URL?"] |
Create a new tab (required before writing) |
tabs |
[] |
List all tabs |
closetab |
["id?"] |
Close a tab |
The Snapshot → @ref Pattern
This is the most powerful browsing pattern. Instead of writing CSS selectors:
- Run
snapshot -ito get an interactive snapshot with labeled elements - The snapshot returns text like:
[Page Title] @e1 [link] "Home" @e2 [button] "Sign In" @e3 [input] "Search..." - Use the
@erefs directly in commands:click @e2,fill @e3 "search query"
This is how the snapshot system works, and it's much more reliable than guessing
CSS selectors. Always snapshot -i first, then use the refs.
Scopes
| Scope | What it allows |
|---|---|
read |
snapshot, text, html, links, screenshot, url, tabs, console, etc. |
write |
goto, click, fill, scroll, newtab, closetab, etc. |
admin |
eval, js, cookies, storage, cookie-import, useragent, etc. |
meta |
tab, diff, frame, responsive, watch |
control |
stop, restart, disconnect, state, handoff — browser-wide destructive ops |
Paired agents get read+write+admin+meta by default; the pairing ceremony is the trust boundary. --restrict narrows the list (it can never grant control). --control adds the control scope (--admin is a legacy alias). Over the tunnel, the js/cookies/storage commands are blocked by the command allowlist regardless of scope; eval works. Pair with --restrict "read,write" when the agent will read untrusted web content — scope caps the prompt-injection blast radius.
To tighten an already-paired agent, re-pair it with the same --client name and the narrower --restrict/--domain: a reducing re-pair revokes the previous session and releases its tabs immediately (the agent must reconnect with the new key), so the old wide access never lingers. Broadening or refreshing keeps the working session with no outage. Re-pairing without --client mints a new agent instead. root is a reserved client name.
Tab Isolation
Each agent owns the tabs it creates. Rules:
- Read: Any agent can read any tab (snapshot, text, screenshot)
- Write: Only the tab owner can write (click, fill, goto, etc.)
- Unowned tabs: Pre-existing tabs are root-only for writes
- First step: Always
newtabbefore trying to interact
Error Codes
| Code | Meaning | What to do |
|---|---|---|
| 401 | Token invalid, expired, or revoked | Ask user to run /pair-agent again |
| 403 | Command not in scope, tab not yours, or not on the tunnel allowlist | Use newtab; the user can re-pair without --restrict or with --control |
| 429 | Rate limit exceeded (>10 req/s) | Wait for Retry-After header |
Security Model
- Physical port separation. Local listener and tunnel listener are separate TCP sockets. ngrok only forwards the tunnel port. Tunnel callers cannot reach bootstrap endpoints at all (404, wrong port).
- Tunnel command allowlist.
/commandover the tunnel only accepts 26 browser-driving commands (goto, click, fill, snapshot, text, newtab, tabs, back, forward, reload, closetab, etc.). Server-management commands (tunnel, pair, token, useragent, js) are denied on the tunnel. - Root token is tunnel-blocked. A request bearing the root token over the tunnel listener returns 403 with a pairing hint. Only scoped session tokens work over the tunnel.
- Setup keys expire in 5 minutes and can only be used once.
- Session tokens expire in 24 hours (configurable).
- The root token never appears in instruction blocks or connection strings.
- Control scope (stop/restart/disconnect) is denied by default and never rides in via a scopes list. Admin is granted at pairing;
js/cookies/storagestay blocked over the tunnel by the command allowlist. Use--restrictfor less-trusted agents. - Tokens can be revoked instantly:
$B tunnel revoke agent-namedeletes the session plus any pending setup keys and verifies against the live agent list.$B tunnel agentsshows who's paired (pending setup keys included).$B stopclears everything — tokens never survive the daemon. - SSE auth uses a 30-minute HttpOnly SameSite=Strict cookie, stream-scope only (never valid against
/command). - Path traversal guarded on
/welcome—GSTACK_SLUGmust match^[a-z0-9_-]+$or falls back to the built-in template. - SSRF guards on
goto,download, and scrape paths — validates URL target against a localhost/private-range blocklist. - Tunnel surface denial logging. Every rejection on the tunnel listener (
path_not_on_tunnel,root_token_on_tunnel,missing_scoped_token,disallowed_command:*) is appended to~/.gstack/security/attempts.jsonlwith timestamp, source IP, path, method. Rate-capped at 60 writes/min. - Egress receipt on tunnel start (v1.63+). Every tunnel session open writes a hash-chained receipt (sink
browse-tunnel) to~/.gstack/security/egress.jsonlBEFORE ngrok forwards anything. Fail-closed: if the receipt can't be written, the tunnel refuses to start. Audit withbin/gstack-egress list/bin/gstack-egress verify. - All agent activity is logged with attribution (clientId).
Known non-goal (tracked as #1136): on Windows, the cookie-import-browser path launches Chrome with --remote-debugging-port=<random>. With App-Bound Encryption v20, a same-user local process can connect to that port and exfiltrate decrypted v20 cookies — an elevation path relative to reading the SQLite DB directly. Fix direction is --remote-debugging-pipe instead of TCP.
Same-Machine Shortcut
If both agents are on the same machine, skip the copy-paste:
$B pair-agent --local openclaw # writes to ~/.openclaw/skills/gstack/browse-remote.json
$B pair-agent --local codex # writes to ~/.codex/skills/gstack/browse-remote.json
$B pair-agent --local cursor # writes to ~/.cursor/skills/gstack/browse-remote.json
No tunnel needed. Uses localhost directly.
ngrok Tunnel Setup
For remote agents on different machines:
- Sign up at ngrok.com (free tier works)
- Copy your auth token from the dashboard
- Save it:
echo 'NGROK_AUTHTOKEN=your_token' > ~/.gstack/ngrok.env - Optionally claim a stable domain:
echo 'NGROK_DOMAIN=your-name.ngrok-free.dev' >> ~/.gstack/ngrok.env - Start with tunnel:
BROWSE_TUNNEL=1 $B restart - Run
$B pair-agent— it will use the tunnel URL automatically