llm-wiki-chrome
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LLM_WIKI_BROWSER_EXECUTOR_NATIVE_SOCKET | No | Exceptional runtime path override for the native socket when the default /tmp directory is inaccessible. Must be set consistently for both the installer and the MCP server. Not part of the normal packaged user flow. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| browser_statusA | Check whether the local connector is online and how many tabs are explicitly shared. |
| browser_tabsA | List only HTTPS tabs the user explicitly shared by clicking the LLM Wiki Browser Executor extension. Call this first; never infer or fabricate a collaboration ID. |
| browser_focusA | Focus one explicitly shared tab without reading or mutating its page. |
| browser_openB | Open and explicitly grant a new tab on the same origin as an existing shared tab. |
| browser_historyC | Go back or forward only when the caller supplies the exact expected same-origin URL. |
| browser_closeC | Close one exact explicitly shared tab through a governed mutation boundary. |
| browser_snapshotB | Read a bounded accessibility-tree projection from one explicitly shared tab. |
| browser_screenshotC | Capture the current viewport of one explicitly shared tab as a private JPEG result. |
| browser_region_screenshotC | Capture one bounded document region from an explicitly shared tab. |
| browser_full_page_screenshotB | Capture a dimension- and byte-capped full-page JPEG from an explicitly shared tab. |
| browser_geometryC | Return bounded private geometry for semantic elements in an explicitly shared tab. |
| browser_waitB | Wait for one bounded semantic condition in an explicitly shared tab. |
| browser_navigateC | Navigate one explicitly shared tab to an exact same-origin HTTPS URL. |
| browser_reloadA | Reload one explicitly shared exact tab, optionally bypassing cache. |
| browser_clickB | Click one uniquely identified accessibility-tree control in an explicitly shared tab. |
| browser_hoverB | Hover one uniquely identified semantic element in an explicitly shared tab. |
| browser_selectC | Open a semantic selector and choose one semantic option. |
| browser_dragC | Drag one semantic element to another using bounded pointer steps. |
| browser_typeC | Focus one semantic control and insert private text in an explicitly shared tab; this does not submit the form. |
| browser_keyC | Dispatch one bounded key chord to an explicitly shared tab. Modifier and Enter/Tab chords can commit or submit a focused form; this crosses the governed mutation boundary. |
| browser_scrollC | Scroll the viewport of one explicitly shared tab by a bounded distance. |
| browser_scroll_toC | Scroll one semantic element into view in an explicitly shared tab. |
| browser_dialogB | Accept or dismiss a browser modal dialog; prompt text stays in a private slot. |
| browser_uploadC | Set one file input from paths inside machine-local registered upload roots. |
| browser_downloadC | Click a semantic download control and verify completed files against registered roots. |
| browser_diagnosticsC | Capture bounded logs, scalar console events, request metadata, and performance metrics. |
| browser_credential_fillA | Focus a field and invoke a local password-manager UI without receiving the secret. |
| browser_record_startC | Start an in-memory typed workflow draft. |
| browser_record_statusC | Return content-free recording state. |
| browser_record_stopA | Stop recording and return a review-required workflow draft. |
| browser_schedule_snapshotC | Schedule one in-memory read-only snapshot while the exact grant and MCP process remain live. |
| browser_schedule_statusA | Return content-free state for in-memory read-only schedules. |
| browser_schedule_cancelC | Cancel one pending in-memory read-only schedule. |
| browser_schedule_resultC | Retrieve and remove one completed private read-only scheduled result. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/nvk/llm-wiki-chrome'
If you have feedback or need assistance with the MCP directory API, please join our Discord server