alvis/AgapHost
1
0
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases Wiki Activity

openclaw-tools: fix browser_invoke tool description (act, not click/type)

Browse Source 
The bridge advertised non-existent gateway actions ("click"/"type"), so Kimi
sent action="type" -> "Unknown action". Corrected to the real vocabulary:
interactions go through action="act" with args={kind, ref, text} (e.g.
{kind:"type", ref:"e59", text:"..."}, {kind:"click", ref}), plus the real
top-level actions (open/navigate/snapshot/screenshot/act/...). Verified Adolf
can now type/click in the browser.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB
This commit is contained in:
alvis
2026-07-16 15:01:18 +00:00
parent 442bc377bf
commit ebb7fb8af3
1 changed files with 14 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
openai/openclaw-tools/server.js
View File

@@ -185,15 +185,21 @@ function buildServer() {
// --- browser --------------------------------------------------------------
server.registerTool('browser_invoke', {
description:
'Generic pass-through to the OpenClaw gateway "browser" agent tool (agent-controlled Chrome/Brave/Edge ' +
'automation: tabs, snapshot, click, type, screenshot). Unlike cron/nodes, "browser" is NOT in the ' +
'gateway HTTP default hard-deny list, so this can work as soon as the gateway is up and its normal ' +
'tools.* policy allows "browser" for the caller — no special P6 HTTP-deny override needed. Args are ' +
'forwarded verbatim as the tool call payload; the exact action/argument vocabulary is defined by the ' +
'running gateway and needs live introspection once it exists (deferred to P6/P7).',
'Drive the OpenClaw gateway "browser" tool (a headless Chromium the agent controls). ' +
'TOP-LEVEL actions (the "action" arg): status, start, stop, profiles, tabs, open, navigate, ' +
'snapshot, screenshot, act, console, dialog, pdf, upload. ' +
'IMPORTANT: there is NO top-level "click"/"type"/"fill"/"evaluate" action — ALL page ' +
'interactions go through action="act" with args={kind, ref, ...}. ' +
'Typical flow: action="open" {url} -> action="snapshot" to get element refs (e.g. "e59") -> ' +
'action="act" to interact. Interaction args examples: ' +
'type text -> {kind:"type", ref:"e59", text:"hello"}; ' +
'click -> {kind:"click", ref:"e33"}; ' +
'fill -> {kind:"fill", ref:"e65", text:"secret"}; ' +
'also kind can be hover|select|press|scrollIntoView|drag|evaluate. ' +
'Refs go stale after navigation/DOM change — if an action reports "ref not found", re-snapshot.',
inputSchema: {
action: z.string().describe('Browser tool action, e.g. "status", "open", "snapshot", "click", "type".'),
args: z.record(z.string(), z.unknown()).optional().describe('Action-specific parameters object.'),
action: z.string().describe('Top-level browser action: open | navigate | snapshot | screenshot | act | tabs | status | start | stop | profiles | console | dialog | pdf | upload. Use "act" for ALL page interactions (click/type/fill), never "click"/"type" directly.'),
args: z.record(z.string(), z.unknown()).optional().describe('Action params. action="open": {url, label?}. action="act": {kind:"type"|"click"|"fill"|"hover"|"select"|..., ref:"eN" from snapshot, text:"..." for type/fill}. action="snapshot": {refs:"aria"} for stable refs.'),
},
}, async ({ action, args }) => invokeGatewayTool('browser', action, args || {}));