{ "schemaVersion": "1.0", "item": { "slug": "midscene-computer-automation", "name": "Midscene Automations Skills for Computer", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/quanru/midscene-computer-automation", "canonicalUrl": "https://clawhub.ai/quanru/midscene-computer-automation", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/midscene-computer-automation", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=midscene-computer-automation", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/midscene-computer-automation" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/midscene-computer-automation", "agentPageUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent", "manifestUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent.json", "briefUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "Desktop Computer Automation", "body": "CRITICAL RULES — VIOLATIONS WILL BREAK THE WORKFLOW:\n\nNever run midscene commands in the background. Each command must run synchronously so you can read its output (especially screenshots) before deciding the next action. Background execution breaks the screenshot-analyze-act loop.\nRun only one midscene command at a time. Wait for the previous command to finish, read the screenshot, then decide the next action. Never chain multiple commands together.\nAllow enough time for each command to complete. Midscene commands involve AI inference and screen interaction, which can take longer than typical shell commands. A typical command needs about 1 minute; complex act commands may need even longer.\nAlways report task results before finishing. After completing the automation task, you MUST proactively summarize the results to the user — including key data found, actions completed, screenshots taken, and any relevant findings. Never silently end after the last automation step; the user expects a complete response in a single interaction.\n\nControl your desktop (macOS, Windows, Linux) using npx @midscene/computer@1. Each CLI command maps directly to an MCP tool — you (the AI agent) act as the brain, deciding which actions to take based on screenshots." }, { "title": "Prerequisites", "body": "Midscene requires models with strong visual grounding capabilities. The following environment variables must be configured — either as system environment variables or in a .env file in the current working directory (Midscene loads .env automatically):\n\nMIDSCENE_MODEL_API_KEY=\"your-api-key\"\nMIDSCENE_MODEL_NAME=\"model-name\"\nMIDSCENE_MODEL_BASE_URL=\"https://...\"\nMIDSCENE_MODEL_FAMILY=\"family-identifier\"\n\nExample: Gemini (Gemini-3-Flash)\n\nMIDSCENE_MODEL_API_KEY=\"your-google-api-key\"\nMIDSCENE_MODEL_NAME=\"gemini-3-flash\"\nMIDSCENE_MODEL_BASE_URL=\"https://generativelanguage.googleapis.com/v1beta/openai/\"\nMIDSCENE_MODEL_FAMILY=\"gemini\"\n\nExample: Qwen 3.5\n\nMIDSCENE_MODEL_API_KEY=\"your-aliyun-api-key\"\nMIDSCENE_MODEL_NAME=\"qwen3.5-plus\"\nMIDSCENE_MODEL_BASE_URL=\"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nMIDSCENE_MODEL_FAMILY=\"qwen3.5\"\nMIDSCENE_MODEL_REASONING_ENABLED=\"false\"\n# If using OpenRouter, set:\n# MIDSCENE_MODEL_API_KEY=\"your-openrouter-api-key\"\n# MIDSCENE_MODEL_NAME=\"qwen/qwen3.5-plus\"\n# MIDSCENE_MODEL_BASE_URL=\"https://openrouter.ai/api/v1\"\n\nExample: Doubao Seed 2.0 Lite\n\nMIDSCENE_MODEL_API_KEY=\"your-doubao-api-key\"\nMIDSCENE_MODEL_NAME=\"doubao-seed-2-0-lite\"\nMIDSCENE_MODEL_BASE_URL=\"https://ark.cn-beijing.volces.com/api/v3\"\nMIDSCENE_MODEL_FAMILY=\"doubao-seed\"\n\nCommonly used models: Doubao Seed 2.0 Lite, Qwen 3.5, Zhipu GLM-4.6V, Gemini-3-Pro, Gemini-3-Flash.\n\nIf the model is not configured, ask the user to set it up. See Model Configuration for supported providers." }, { "title": "Connect to Desktop", "body": "npx @midscene/computer@1 connect\nnpx @midscene/computer@1 connect --displayId " }, { "title": "List Displays", "body": "npx @midscene/computer@1 list_displays" }, { "title": "Take Screenshot", "body": "npx @midscene/computer@1 take_screenshot\n\nAfter taking a screenshot, read the saved image file to understand the current screen state before deciding the next action." }, { "title": "Perform Action", "body": "Use act to interact with the computer and get the result. It autonomously handles all UI interactions internally — clicking, typing, scrolling, waiting, and navigating — so you should give it complex, high-level tasks as a whole rather than breaking them into small steps. Describe what you want to do and the desired effect in natural language:\n\n# specific instructions\nnpx @midscene/computer@1 act --prompt \"type hello world in the search field and press Enter\"\nnpx @midscene/computer@1 act --prompt \"drag the file icon to the Trash\"\n\n# or target-driven instructions\nnpx @midscene/computer@1 act --prompt \"search for the weather in Shanghai using the Chrome browser, tell me the result\"" }, { "title": "Disconnect", "body": "npx @midscene/computer@1 disconnect" }, { "title": "Workflow Pattern", "body": "Since CLI commands are stateless between invocations, follow this pattern:\n\nConnect to establish a session\nHealth check — observe the output of the connect command. If connect already performed a health check (screenshot and mouse movement test), no additional check is needed. If connect did not perform a health check, do one manually: take a screenshot and verify it succeeds, then move the mouse to a random position (act --prompt \"move the mouse to a random position\") and verify it succeeds. If either step fails, stop and troubleshoot before continuing. Only proceed to the next steps after both checks pass without errors.\nLaunch the target app and take screenshot to see the current state, make sure the app is launched and visible on the screen.\nExecute action using act to perform the desired action or target-driven instructions.\nDisconnect when done\nReport results — summarize what was accomplished, present key findings and data extracted during the task, and list any generated files (screenshots, logs, etc.) with their paths" }, { "title": "Best Practices", "body": "Always run a health check first: After connecting, observe the output of the connect command. If connect already performed a health check (screenshot and mouse movement test), no additional check is needed. If it did not, do one manually: take a screenshot and move the mouse to a random position. Both must succeed (no errors) before proceeding with any further operations. This catches environment issues early.\nBring the target app to the foreground before using this skill: For best efficiency, bring the app to the foreground using other means (e.g., open -a on macOS, start on Windows) before invoking any midscene commands. Then take a screenshot to confirm the app is actually in the foreground. Only after visual confirmation should you proceed with UI automation using this skill. Avoid using Spotlight, Start menu search, or other launcher-based approaches through midscene — they involve transient UI, multiple AI inference steps, and are significantly slower.\nBe specific about UI elements: Instead of vague descriptions, provide clear, specific details. Say \"the red close button in the top-left corner of the Safari window\" instead of \"the close button\".\nDescribe locations when possible: Help target elements by describing their position (e.g., \"the icon in the top-right corner of the menu bar\", \"the third item in the left sidebar\").\nNever run in background: Every midscene command must run synchronously — background execution breaks the screenshot-analyze-act loop.\nCheck for multiple displays: If you launched an app but cannot see it on the screenshot, the app window may have opened on a different display. Use list_displays to check available displays. You have two options: either move the app window to the current display, or use connect --displayId to switch to the display where the app is.\nBatch related operations into a single act command: When performing consecutive operations within the same app, combine them into one act prompt instead of splitting them into separate commands. For example, \"search for X, click the first result, and scroll down to see more details\" should be a single act call, not three. This reduces round-trips, avoids unnecessary screenshot-analyze cycles, and is significantly faster.\nSet up PATH before running (macOS): On macOS, some commands (e.g., system_profiler) may not be found if the PATH is incomplete. Before running any midscene commands, ensure the PATH includes the standard system directories:\nexport PATH=\"/usr/sbin:/usr/bin:/bin:/sbin:$PATH\"\n\nThis prevents screenshot failures caused by missing system utilities.\nAlways report results after completion: After finishing the automation task, you MUST proactively present the results to the user without waiting for them to ask. This includes: (1) the answer to the user's original question or the outcome of the requested task, (2) key data extracted or observed during execution, (3) screenshots and other generated files with their paths, (4) a brief summary of steps taken. Do NOT silently finish after the last automation command — the user expects complete results in a single interaction.\n\nExample — Context menu interaction:\n\nnpx @midscene/computer@1 act --prompt \"right-click the file icon and select Delete from the context menu\"\nnpx @midscene/computer@1 take_screenshot\n\nExample — Dropdown menu:\n\nnpx @midscene/computer@1 act --prompt \"open the File menu and click New Window\"\nnpx @midscene/computer@1 take_screenshot" }, { "title": "macOS: Accessibility Permission Denied", "body": "Your terminal app does not have Accessibility access:\n\nOpen System Settings > Privacy & Security > Accessibility\nAdd your terminal app and enable it\nRestart your terminal app after granting permission" }, { "title": "macOS: Xcode Command Line Tools Not Found", "body": "xcode-select --install" }, { "title": "API Key Not Set", "body": "Check .env file contains MIDSCENE_MODEL_API_KEY=." }, { "title": "macOS: Screenshot Fails with system_profiler Not Found", "body": "If take_screenshot fails with an error like system_profiler: command not found, the PATH environment variable is likely incomplete. Fix it by running:\n\nexport PATH=\"/usr/sbin:/usr/bin:/bin:/sbin:$PATH\"\n\nThen retry the screenshot command." }, { "title": "AI Cannot Find the Element", "body": "Take a screenshot to verify the element is actually visible\nUse more specific descriptions (include color, position, surrounding text)\nEnsure the element is not hidden behind another window" } ], "body": "Desktop Computer Automation\n\nCRITICAL RULES — VIOLATIONS WILL BREAK THE WORKFLOW:\n\nNever run midscene commands in the background. Each command must run synchronously so you can read its output (especially screenshots) before deciding the next action. Background execution breaks the screenshot-analyze-act loop.\nRun only one midscene command at a time. Wait for the previous command to finish, read the screenshot, then decide the next action. Never chain multiple commands together.\nAllow enough time for each command to complete. Midscene commands involve AI inference and screen interaction, which can take longer than typical shell commands. A typical command needs about 1 minute; complex act commands may need even longer.\nAlways report task results before finishing. After completing the automation task, you MUST proactively summarize the results to the user — including key data found, actions completed, screenshots taken, and any relevant findings. Never silently end after the last automation step; the user expects a complete response in a single interaction.\n\nControl your desktop (macOS, Windows, Linux) using npx @midscene/computer@1. Each CLI command maps directly to an MCP tool — you (the AI agent) act as the brain, deciding which actions to take based on screenshots.\n\nPrerequisites\n\nMidscene requires models with strong visual grounding capabilities. The following environment variables must be configured — either as system environment variables or in a .env file in the current working directory (Midscene loads .env automatically):\n\nMIDSCENE_MODEL_API_KEY=\"your-api-key\"\nMIDSCENE_MODEL_NAME=\"model-name\"\nMIDSCENE_MODEL_BASE_URL=\"https://...\"\nMIDSCENE_MODEL_FAMILY=\"family-identifier\"\n\n\nExample: Gemini (Gemini-3-Flash)\n\nMIDSCENE_MODEL_API_KEY=\"your-google-api-key\"\nMIDSCENE_MODEL_NAME=\"gemini-3-flash\"\nMIDSCENE_MODEL_BASE_URL=\"https://generativelanguage.googleapis.com/v1beta/openai/\"\nMIDSCENE_MODEL_FAMILY=\"gemini\"\n\n\nExample: Qwen 3.5\n\nMIDSCENE_MODEL_API_KEY=\"your-aliyun-api-key\"\nMIDSCENE_MODEL_NAME=\"qwen3.5-plus\"\nMIDSCENE_MODEL_BASE_URL=\"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nMIDSCENE_MODEL_FAMILY=\"qwen3.5\"\nMIDSCENE_MODEL_REASONING_ENABLED=\"false\"\n# If using OpenRouter, set:\n# MIDSCENE_MODEL_API_KEY=\"your-openrouter-api-key\"\n# MIDSCENE_MODEL_NAME=\"qwen/qwen3.5-plus\"\n# MIDSCENE_MODEL_BASE_URL=\"https://openrouter.ai/api/v1\"\n\n\nExample: Doubao Seed 2.0 Lite\n\nMIDSCENE_MODEL_API_KEY=\"your-doubao-api-key\"\nMIDSCENE_MODEL_NAME=\"doubao-seed-2-0-lite\"\nMIDSCENE_MODEL_BASE_URL=\"https://ark.cn-beijing.volces.com/api/v3\"\nMIDSCENE_MODEL_FAMILY=\"doubao-seed\"\n\n\nCommonly used models: Doubao Seed 2.0 Lite, Qwen 3.5, Zhipu GLM-4.6V, Gemini-3-Pro, Gemini-3-Flash.\n\nIf the model is not configured, ask the user to set it up. See Model Configuration for supported providers.\n\nCommands\nConnect to Desktop\nnpx @midscene/computer@1 connect\nnpx @midscene/computer@1 connect --displayId \n\nList Displays\nnpx @midscene/computer@1 list_displays\n\nTake Screenshot\nnpx @midscene/computer@1 take_screenshot\n\n\nAfter taking a screenshot, read the saved image file to understand the current screen state before deciding the next action.\n\nPerform Action\n\nUse act to interact with the computer and get the result. It autonomously handles all UI interactions internally — clicking, typing, scrolling, waiting, and navigating — so you should give it complex, high-level tasks as a whole rather than breaking them into small steps. Describe what you want to do and the desired effect in natural language:\n\n# specific instructions\nnpx @midscene/computer@1 act --prompt \"type hello world in the search field and press Enter\"\nnpx @midscene/computer@1 act --prompt \"drag the file icon to the Trash\"\n\n# or target-driven instructions\nnpx @midscene/computer@1 act --prompt \"search for the weather in Shanghai using the Chrome browser, tell me the result\"\n\nDisconnect\nnpx @midscene/computer@1 disconnect\n\nWorkflow Pattern\n\nSince CLI commands are stateless between invocations, follow this pattern:\n\nConnect to establish a session\nHealth check — observe the output of the connect command. If connect already performed a health check (screenshot and mouse movement test), no additional check is needed. If connect did not perform a health check, do one manually: take a screenshot and verify it succeeds, then move the mouse to a random position (act --prompt \"move the mouse to a random position\") and verify it succeeds. If either step fails, stop and troubleshoot before continuing. Only proceed to the next steps after both checks pass without errors.\nLaunch the target app and take screenshot to see the current state, make sure the app is launched and visible on the screen.\nExecute action using act to perform the desired action or target-driven instructions.\nDisconnect when done\nReport results — summarize what was accomplished, present key findings and data extracted during the task, and list any generated files (screenshots, logs, etc.) with their paths\nBest Practices\nAlways run a health check first: After connecting, observe the output of the connect command. If connect already performed a health check (screenshot and mouse movement test), no additional check is needed. If it did not, do one manually: take a screenshot and move the mouse to a random position. Both must succeed (no errors) before proceeding with any further operations. This catches environment issues early.\nBring the target app to the foreground before using this skill: For best efficiency, bring the app to the foreground using other means (e.g., open -a on macOS, start on Windows) before invoking any midscene commands. Then take a screenshot to confirm the app is actually in the foreground. Only after visual confirmation should you proceed with UI automation using this skill. Avoid using Spotlight, Start menu search, or other launcher-based approaches through midscene — they involve transient UI, multiple AI inference steps, and are significantly slower.\nBe specific about UI elements: Instead of vague descriptions, provide clear, specific details. Say \"the red close button in the top-left corner of the Safari window\" instead of \"the close button\".\nDescribe locations when possible: Help target elements by describing their position (e.g., \"the icon in the top-right corner of the menu bar\", \"the third item in the left sidebar\").\nNever run in background: Every midscene command must run synchronously — background execution breaks the screenshot-analyze-act loop.\nCheck for multiple displays: If you launched an app but cannot see it on the screenshot, the app window may have opened on a different display. Use list_displays to check available displays. You have two options: either move the app window to the current display, or use connect --displayId to switch to the display where the app is.\nBatch related operations into a single act command: When performing consecutive operations within the same app, combine them into one act prompt instead of splitting them into separate commands. For example, \"search for X, click the first result, and scroll down to see more details\" should be a single act call, not three. This reduces round-trips, avoids unnecessary screenshot-analyze cycles, and is significantly faster.\nSet up PATH before running (macOS): On macOS, some commands (e.g., system_profiler) may not be found if the PATH is incomplete. Before running any midscene commands, ensure the PATH includes the standard system directories:\nexport PATH=\"/usr/sbin:/usr/bin:/bin:/sbin:$PATH\"\n\nThis prevents screenshot failures caused by missing system utilities.\nAlways report results after completion: After finishing the automation task, you MUST proactively present the results to the user without waiting for them to ask. This includes: (1) the answer to the user's original question or the outcome of the requested task, (2) key data extracted or observed during execution, (3) screenshots and other generated files with their paths, (4) a brief summary of steps taken. Do NOT silently finish after the last automation command — the user expects complete results in a single interaction.\n\nExample — Context menu interaction:\n\nnpx @midscene/computer@1 act --prompt \"right-click the file icon and select Delete from the context menu\"\nnpx @midscene/computer@1 take_screenshot\n\n\nExample — Dropdown menu:\n\nnpx @midscene/computer@1 act --prompt \"open the File menu and click New Window\"\nnpx @midscene/computer@1 take_screenshot\n\nTroubleshooting\nmacOS: Accessibility Permission Denied\n\nYour terminal app does not have Accessibility access:\n\nOpen System Settings > Privacy & Security > Accessibility\nAdd your terminal app and enable it\nRestart your terminal app after granting permission\nmacOS: Xcode Command Line Tools Not Found\nxcode-select --install\n\nAPI Key Not Set\n\nCheck .env file contains MIDSCENE_MODEL_API_KEY=.\n\nmacOS: Screenshot Fails with system_profiler Not Found\n\nIf take_screenshot fails with an error like system_profiler: command not found, the PATH environment variable is likely incomplete. Fix it by running:\n\nexport PATH=\"/usr/sbin:/usr/bin:/bin:/sbin:$PATH\"\n\n\nThen retry the screenshot command.\n\nAI Cannot Find the Element\nTake a screenshot to verify the element is actually visible\nUse more specific descriptions (include color, position, surrounding text)\nEnsure the element is not hidden behind another window" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/quanru/midscene-computer-automation", "publisherUrl": "https://clawhub.ai/quanru/midscene-computer-automation", "owner": "quanru", "version": "1.0.3", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/midscene-computer-automation", "downloadUrl": "https://openagent3.xyz/downloads/midscene-computer-automation", "agentUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent", "manifestUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent.json", "briefUrl": "https://openagent3.xyz/skills/midscene-computer-automation/agent.md" } }