computer-mcp
by FloraAurella
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| COMPUTER_RUNTIME_BIN | Yes | Absolute path to the compiled computer-runtime binary. | |
| COMPUTER_RUNTIME_ARGS | No | Additional arguments for the runtime, e.g. --mock to use the built-in mock platform. | |
| COMPUTER_TOOL_PROFILE | No | Static tool surface: 'desktop-core', 'desktop-vision', 'browser', 'full', or 'debug' (default 'full'). | full |
| COMPUTER_VISION_MODEL | No | Model name for the 'openai-compatible' vision provider. | |
| COMPUTER_BROWSER_HEADED | No | Set to '1' to launch browser.* tools with a headed browser (default is headless). | 0 |
| COMPUTER_CAPTURE_TTL_MS | No | Capture byte TTL in milliseconds (default 600000). | 600000 |
| COMPUTER_VISION_API_KEY | No | Optional API key for the 'openai-compatible' vision provider. | |
| COMPUTER_VISION_ENABLED | No | Set to '1' to enable vision grounding for computer_visual_observe/computer_resolve (default off). | 0 |
| COMPUTER_CONTEXT_METRICS | No | Set to '1' to log schema character counts, Tool Result byte counts, element counts and vision attempts to stderr. | 0 |
| COMPUTER_ENABLE_EVALUATE | No | Set to '0' to disable the browser_evaluate tool (security hardening). | 1 |
| COMPUTER_SCREENSHOT_MODE | No | Screenshot return semantics: 'reference', 'image', or 'auto' (default 'reference'). | reference |
| COMPUTER_VISION_BASE_URL | No | API root URL for the 'openai-compatible' vision provider. | |
| COMPUTER_VISION_PROVIDER | No | Vision provider: 'mock', 'remote', or 'openai-compatible' (default 'mock'). | mock |
| COMPUTER_CAPTURE_MAX_EDGE | No | Longest edge for preview images (default 1568). | 1568 |
| COMPUTER_LEGACY_SCREENSHOT | No | Short-term transition flag; same as 'image', with deprecation warning. | 0 |
| COMPUTER_VISION_MAX_RETRIES | No | Maximum retries for managed vision (default 1). | 1 |
| COMPUTER_VISION_REMOTE_UPLOAD | No | Second switch for all remote providers. If unset, screenshots will never be uploaded to external endpoints. | 0 |
| COMPUTER_VISION_MAX_IMAGE_BYTES | No | Maximum image size in bytes for vision (default 8MiB). | 8388608 |
| COMPUTER_ALLOW_RAW_IMAGE_CONTENT | No | Set to '1' to explicitly allow computer_screenshot_image to return ImageContent (default disabled). | 0 |
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
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
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/FloraAurella/computer-use-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server