{
  "schemaVersion": "1.0",
  "item": {
    "slug": "cursor-ide-agent",
    "name": "Cursor IDE Agent",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/xiaoyaner0201/cursor-ide-agent",
    "canonicalUrl": "https://clawhub.ai/xiaoyaner0201/cursor-ide-agent",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/cursor-ide-agent",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=cursor-ide-agent",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "README.md",
      "SKILL.md",
      "_meta.json"
    ],
    "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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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/cursor-ide-agent"
    },
    "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/cursor-ide-agent",
    "agentPageUrl": "https://openagent3.xyz/skills/cursor-ide-agent/agent",
    "manifestUrl": "https://openagent3.xyz/skills/cursor-ide-agent/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/cursor-ide-agent/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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Cursor Agent Skill",
        "body": "Two ways to use Cursor Agent from OpenClaw, for different scenarios."
      },
      {
        "title": "Related",
        "body": "OpenClaw Node for VS Code — VS Code/Cursor extension that enables the Node path (install this first for Path 2)\nvscode-node — Standalone skill for the Node path only (if you don't need Cursor CLI)\nSource: github.com/xiaoyaner-home/openclaw-vscode"
      },
      {
        "title": "Path Selection",
        "body": "ScenarioPathWhyQuick coding task, bug fix, refactorCLIFast, no setup, works anywhereGenerate code, review PR, write testsCLINon-interactive -p mode is perfectFix type errors using real diagnosticsNodediagnostics.get shows actual TS/lint errorsNavigate definitions/references firstNodelang.definition, lang.referencesRun project tests and iterateNodetest.run + test.results loopDebug with breakpointsNodeFull debug protocolTargeted changes to a specific projectNodeIDE workspace context is precise\n\nDefault: CLI. Use Node only when you specifically need IDE intelligence."
      },
      {
        "title": "Prerequisites",
        "body": "# Install\ncurl https://cursor.com/install -fsS | bash\n\n# Login\nagent login\n\n# Verify\nagent --version"
      },
      {
        "title": "Modes",
        "body": "ModeFlagUse CaseAgent(default)Full coding — reads, writes, runs commandsPlan--plan or --mode=planDesign approach first, then choose local or cloud executionAsk--mode=askRead-only codebase exploration, no edits"
      },
      {
        "title": "Interactive Mode",
        "body": "# Start interactive session\nagent\n\n# Start with prompt\nagent \"refactor the auth module to use JWT tokens\"\n\n# Start in plan mode\nagent --plan \"design a caching layer for the API\"\n\n# Start in ask mode\nagent --mode=ask \"explain how the auth middleware works\""
      },
      {
        "title": "Non-Interactive Mode (Automation)",
        "body": "# One-shot task (prints result, exits)\nagent -p \"find and fix all unused imports in src/\"\n\n# With specific model\nagent -p \"review this code for security issues\" --model gpt-5.2\n\n# JSON output for parsing\nagent -p \"list all TODO comments\" --output-format json\n\n# Streaming JSON (real-time)\nagent -p \"run tests and report\" --output-format stream-json --stream-partial-output\n\n# Force mode (auto-apply changes, no confirmation)\nagent -p \"fix all linting errors\" --force"
      },
      {
        "title": "Cloud Agent Handoff",
        "body": "Push work to Cursor's cloud to continue running while you're away:\n\n# Start directly in cloud\nagent -c \"refactor the auth module and add comprehensive tests\"\n\n# Mid-conversation: prepend & to send to cloud\n& refactor the auth module and add comprehensive tests\n\nPick up at cursor.com/agents."
      },
      {
        "title": "Session Management",
        "body": "agent ls              # List previous conversations\nagent resume          # Resume most recent\nagent --continue      # Continue previous session\nagent --resume=\"id\"   # Resume specific conversation"
      },
      {
        "title": "Slash Commands (Interactive)",
        "body": "CommandAction/planSwitch to Plan mode / view current plan/askSwitch to Ask mode/modelsSwitch AI model/compressSummarize conversation, free context/rulesCreate/edit rules/commandsCreate/edit custom commands/mcp enable <name>Enable MCP server/mcp disable <name>Disable MCP server/sandboxConfigure sandbox mode/max-mode [on|off]Toggle Max Mode/resumeResume previous conversation"
      },
      {
        "title": "Keyboard Shortcuts",
        "body": "ShortcutActionShift+TabRotate modes (Agent → Plan → Ask)Shift+EnterInsert newline (multi-line prompt)Ctrl+RReview changes (i for instructions, arrows to navigate)Ctrl+DExit (double-press for safety)ArrowUpCycle previous messages"
      },
      {
        "title": "Context & Rules",
        "body": "The CLI automatically loads:\n\n.cursor/rules directory\nAGENTS.md at project root\nCLAUDE.md at project root\nMCP servers from mcp.json\n\nUse @filename or @directory/ in interactive mode to include context."
      },
      {
        "title": "⚠️ Using CLI from OpenClaw (PTY Required)",
        "body": "Cursor CLI is an interactive TUI — it needs a real terminal. Use pty:true:\n\n# ✅ Correct — with PTY\nexec pty:true command:\"agent -p 'Your task'\" workdir:/path/to/project\n\n# ✅ Background for longer tasks\nexec pty:true background:true command:\"agent -p 'Build REST API'\" workdir:/path/to/project\n\n# ❌ Wrong — will hang\nexec command:\"agent -p 'Your task'\"\n\nFor long tasks, use background + poll:\n\n# Start\nexec pty:true background:true workdir:~/project command:\"agent -p 'Add comprehensive tests for the auth module' --force\"\n\n# Check progress\nprocess action:log sessionId:XXX\n\n# Check if done\nprocess action:poll sessionId:XXX"
      },
      {
        "title": "Sandbox Controls",
        "body": "# Start with sandbox enabled\nagent --sandbox enabled\n\n# Start with sandbox disabled\nagent --sandbox disabled\n\n# Configure interactively\n/sandbox\n\nSandbox supports granular network access controls — define which domains the agent can reach."
      },
      {
        "title": "Path 2: VS Code / Cursor Node",
        "body": "Remote-control a Cursor/VS Code IDE through the OpenClaw Node protocol. The IDE must have the openclaw-node-vscode extension installed and connected."
      },
      {
        "title": "Prerequisites",
        "body": "Extension installed: VS Code Marketplace\nNode visible in nodes status\nExtension status bar shows 🟢"
      },
      {
        "title": "Invocation Pattern",
        "body": "nodes invoke --node \"<name>\" --invokeCommand \"<cmd>\" --invokeParamsJson '{\"key\":\"val\"}'"
      },
      {
        "title": "Timeout Guide",
        "body": "OperationinvokeTimeoutMsNotesFile/editor/lang15000Fast IDE operationsGit30000May involve disk I/OTest60000Depends on test suiteAgent plan/ask180000AI thinking timeAgent run300000Full coding task"
      },
      {
        "title": "Command Reference",
        "body": "CategoryPrefixKey CommandsFilevscode.file.*read, write, edit, deleteDirectoryvscode.dir.*listLanguagevscode.lang.*definition, references, hover, symbols, rename, codeActions, formatEditorvscode.editor.*context, openFiles, selectionsDiagnosticsvscode.diagnostics.*get (errors/warnings)Gitvscode.git.*status, diff, log, blame, stage, unstage, commit, stashTestvscode.test.*list, run, resultsDebugvscode.debug.*launch, stop, breakpoint, evaluate, stackTrace, variables, statusAgentvscode.agent.*status, run, setupWorkspacevscode.workspace.*info"
      },
      {
        "title": "Quick Examples",
        "body": "# Read a file\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.file.read\" \\\n  --invokeParamsJson '{\"path\":\"src/main.ts\"}'\n\n# Get diagnostics (real type errors!)\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.diagnostics.get\"\n\n# Go to definition\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.lang.definition\" \\\n  --invokeParamsJson '{\"path\":\"src/main.ts\",\"line\":10,\"character\":5}'\n\n# Git status + commit\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.status\"\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.stage\" \\\n  --invokeParamsJson '{\"paths\":[\"src/main.ts\"]}'\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.commit\" \\\n  --invokeParamsJson '{\"message\":\"fix: resolve type error\"}'\n\n# Delegate to Cursor Agent (through IDE)\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.agent.run\" \\\n  --invokeParamsJson '{\"prompt\":\"Add error handling to all API endpoints\",\"mode\":\"plan\"}' \\\n  --invokeTimeoutMs 180000"
      },
      {
        "title": "Node Workflow: Fix → Verify → Commit",
        "body": "The real power of Node path — a closed loop with IDE intelligence:\n\n1. diagnostics.get           → Find real errors\n2. vscode.agent.run (fix)    → Let Cursor Agent fix them\n3. diagnostics.get           → Verify errors resolved\n4. test.run                  → Run tests\n5. test.results              → Check results\n6. git.diff                  → Review changes\n7. git.stage + git.commit    → Ship it\n\nNo tmux, no TTY hacks — all through VS Code API."
      },
      {
        "title": "Combined Workflow Example",
        "body": "Use CLI for broad tasks, switch to Node for precision:\n\n1. CLI: agent -p \"implement user authentication module\" --force\n   → Generates the initial code quickly\n\n2. Node: vscode.diagnostics.get\n   → Reveals 3 type errors the CLI missed\n\n3. Node: vscode.agent.run '{\"prompt\":\"fix these type errors: ...\"}'\n   → Fixes with full IDE context\n\n4. Node: vscode.test.run\n   → Runs test suite\n\n5. Node: vscode.git.stage + vscode.git.commit\n   → Clean commit"
      },
      {
        "title": "Error Handling",
        "body": "ErrorCauseFixCLI hangsNo PTYAdd pty:true to execnode not foundExtension disconnectedCheck VS Code status barcommand not allowedGateway whitelistAdd to gateway.nodes.allowCommandstimeoutOperation too longIncrease invokeTimeoutMspath traversal blockedAbsolute path usedUse relative paths for Node"
      },
      {
        "title": "Security",
        "body": "CLI: Respects sandbox mode, command approval, rules\nNode: All paths relative to workspace, Ed25519 device identity, Gateway approval required\nBoth: No raw shell access by default"
      }
    ],
    "body": "Cursor Agent Skill\n\nTwo ways to use Cursor Agent from OpenClaw, for different scenarios.\n\nRelated\nOpenClaw Node for VS Code — VS Code/Cursor extension that enables the Node path (install this first for Path 2)\nvscode-node — Standalone skill for the Node path only (if you don't need Cursor CLI)\nSource: github.com/xiaoyaner-home/openclaw-vscode\nPath Selection\nScenario\tPath\tWhy\nQuick coding task, bug fix, refactor\tCLI\tFast, no setup, works anywhere\nGenerate code, review PR, write tests\tCLI\tNon-interactive -p mode is perfect\nFix type errors using real diagnostics\tNode\tdiagnostics.get shows actual TS/lint errors\nNavigate definitions/references first\tNode\tlang.definition, lang.references\nRun project tests and iterate\tNode\ttest.run + test.results loop\nDebug with breakpoints\tNode\tFull debug protocol\nTargeted changes to a specific project\tNode\tIDE workspace context is precise\n\nDefault: CLI. Use Node only when you specifically need IDE intelligence.\n\nPath 1: CLI (Local Cursor Agent)\nPrerequisites\n# Install\ncurl https://cursor.com/install -fsS | bash\n\n# Login\nagent login\n\n# Verify\nagent --version\n\nModes\nMode\tFlag\tUse Case\nAgent\t(default)\tFull coding — reads, writes, runs commands\nPlan\t--plan or --mode=plan\tDesign approach first, then choose local or cloud execution\nAsk\t--mode=ask\tRead-only codebase exploration, no edits\nInteractive Mode\n# Start interactive session\nagent\n\n# Start with prompt\nagent \"refactor the auth module to use JWT tokens\"\n\n# Start in plan mode\nagent --plan \"design a caching layer for the API\"\n\n# Start in ask mode\nagent --mode=ask \"explain how the auth middleware works\"\n\nNon-Interactive Mode (Automation)\n# One-shot task (prints result, exits)\nagent -p \"find and fix all unused imports in src/\"\n\n# With specific model\nagent -p \"review this code for security issues\" --model gpt-5.2\n\n# JSON output for parsing\nagent -p \"list all TODO comments\" --output-format json\n\n# Streaming JSON (real-time)\nagent -p \"run tests and report\" --output-format stream-json --stream-partial-output\n\n# Force mode (auto-apply changes, no confirmation)\nagent -p \"fix all linting errors\" --force\n\nCloud Agent Handoff\n\nPush work to Cursor's cloud to continue running while you're away:\n\n# Start directly in cloud\nagent -c \"refactor the auth module and add comprehensive tests\"\n\n# Mid-conversation: prepend & to send to cloud\n& refactor the auth module and add comprehensive tests\n\n\nPick up at cursor.com/agents.\n\nSession Management\nagent ls              # List previous conversations\nagent resume          # Resume most recent\nagent --continue      # Continue previous session\nagent --resume=\"id\"   # Resume specific conversation\n\nSlash Commands (Interactive)\nCommand\tAction\n/plan\tSwitch to Plan mode / view current plan\n/ask\tSwitch to Ask mode\n/models\tSwitch AI model\n/compress\tSummarize conversation, free context\n/rules\tCreate/edit rules\n/commands\tCreate/edit custom commands\n/mcp enable <name>\tEnable MCP server\n/mcp disable <name>\tDisable MCP server\n/sandbox\tConfigure sandbox mode\n/max-mode [on|off]\tToggle Max Mode\n/resume\tResume previous conversation\nKeyboard Shortcuts\nShortcut\tAction\nShift+Tab\tRotate modes (Agent → Plan → Ask)\nShift+Enter\tInsert newline (multi-line prompt)\nCtrl+R\tReview changes (i for instructions, arrows to navigate)\nCtrl+D\tExit (double-press for safety)\nArrowUp\tCycle previous messages\nContext & Rules\n\nThe CLI automatically loads:\n\n.cursor/rules directory\nAGENTS.md at project root\nCLAUDE.md at project root\nMCP servers from mcp.json\n\nUse @filename or @directory/ in interactive mode to include context.\n\n⚠️ Using CLI from OpenClaw (PTY Required)\n\nCursor CLI is an interactive TUI — it needs a real terminal. Use pty:true:\n\n# ✅ Correct — with PTY\nexec pty:true command:\"agent -p 'Your task'\" workdir:/path/to/project\n\n# ✅ Background for longer tasks\nexec pty:true background:true command:\"agent -p 'Build REST API'\" workdir:/path/to/project\n\n# ❌ Wrong — will hang\nexec command:\"agent -p 'Your task'\"\n\n\nFor long tasks, use background + poll:\n\n# Start\nexec pty:true background:true workdir:~/project command:\"agent -p 'Add comprehensive tests for the auth module' --force\"\n\n# Check progress\nprocess action:log sessionId:XXX\n\n# Check if done\nprocess action:poll sessionId:XXX\n\nSandbox Controls\n# Start with sandbox enabled\nagent --sandbox enabled\n\n# Start with sandbox disabled\nagent --sandbox disabled\n\n# Configure interactively\n/sandbox\n\n\nSandbox supports granular network access controls — define which domains the agent can reach.\n\nPath 2: VS Code / Cursor Node\n\nRemote-control a Cursor/VS Code IDE through the OpenClaw Node protocol. The IDE must have the openclaw-node-vscode extension installed and connected.\n\nPrerequisites\nExtension installed: VS Code Marketplace\nNode visible in nodes status\nExtension status bar shows 🟢\nInvocation Pattern\nnodes invoke --node \"<name>\" --invokeCommand \"<cmd>\" --invokeParamsJson '{\"key\":\"val\"}'\n\nTimeout Guide\nOperation\tinvokeTimeoutMs\tNotes\nFile/editor/lang\t15000\tFast IDE operations\nGit\t30000\tMay involve disk I/O\nTest\t60000\tDepends on test suite\nAgent plan/ask\t180000\tAI thinking time\nAgent run\t300000\tFull coding task\nCommand Reference\nCategory\tPrefix\tKey Commands\nFile\tvscode.file.*\tread, write, edit, delete\nDirectory\tvscode.dir.*\tlist\nLanguage\tvscode.lang.*\tdefinition, references, hover, symbols, rename, codeActions, format\nEditor\tvscode.editor.*\tcontext, openFiles, selections\nDiagnostics\tvscode.diagnostics.*\tget (errors/warnings)\nGit\tvscode.git.*\tstatus, diff, log, blame, stage, unstage, commit, stash\nTest\tvscode.test.*\tlist, run, results\nDebug\tvscode.debug.*\tlaunch, stop, breakpoint, evaluate, stackTrace, variables, status\nAgent\tvscode.agent.*\tstatus, run, setup\nWorkspace\tvscode.workspace.*\tinfo\nQuick Examples\n# Read a file\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.file.read\" \\\n  --invokeParamsJson '{\"path\":\"src/main.ts\"}'\n\n# Get diagnostics (real type errors!)\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.diagnostics.get\"\n\n# Go to definition\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.lang.definition\" \\\n  --invokeParamsJson '{\"path\":\"src/main.ts\",\"line\":10,\"character\":5}'\n\n# Git status + commit\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.status\"\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.stage\" \\\n  --invokeParamsJson '{\"paths\":[\"src/main.ts\"]}'\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.git.commit\" \\\n  --invokeParamsJson '{\"message\":\"fix: resolve type error\"}'\n\n# Delegate to Cursor Agent (through IDE)\nnodes invoke --node \"my-cursor\" --invokeCommand \"vscode.agent.run\" \\\n  --invokeParamsJson '{\"prompt\":\"Add error handling to all API endpoints\",\"mode\":\"plan\"}' \\\n  --invokeTimeoutMs 180000\n\nNode Workflow: Fix → Verify → Commit\n\nThe real power of Node path — a closed loop with IDE intelligence:\n\n1. diagnostics.get           → Find real errors\n2. vscode.agent.run (fix)    → Let Cursor Agent fix them\n3. diagnostics.get           → Verify errors resolved\n4. test.run                  → Run tests\n5. test.results              → Check results\n6. git.diff                  → Review changes\n7. git.stage + git.commit    → Ship it\n\n\nNo tmux, no TTY hacks — all through VS Code API.\n\nCombined Workflow Example\n\nUse CLI for broad tasks, switch to Node for precision:\n\n1. CLI: agent -p \"implement user authentication module\" --force\n   → Generates the initial code quickly\n\n2. Node: vscode.diagnostics.get\n   → Reveals 3 type errors the CLI missed\n\n3. Node: vscode.agent.run '{\"prompt\":\"fix these type errors: ...\"}'\n   → Fixes with full IDE context\n\n4. Node: vscode.test.run\n   → Runs test suite\n\n5. Node: vscode.git.stage + vscode.git.commit\n   → Clean commit\n\nError Handling\nError\tCause\tFix\nCLI hangs\tNo PTY\tAdd pty:true to exec\nnode not found\tExtension disconnected\tCheck VS Code status bar\ncommand not allowed\tGateway whitelist\tAdd to gateway.nodes.allowCommands\ntimeout\tOperation too long\tIncrease invokeTimeoutMs\npath traversal blocked\tAbsolute path used\tUse relative paths for Node\nSecurity\nCLI: Respects sandbox mode, command approval, rules\nNode: All paths relative to workspace, Ed25519 device identity, Gateway approval required\nBoth: No raw shell access by default"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/xiaoyaner0201/cursor-ide-agent",
    "publisherUrl": "https://clawhub.ai/xiaoyaner0201/cursor-ide-agent",
    "owner": "xiaoyaner0201",
    "version": "3.0.2",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/cursor-ide-agent",
    "downloadUrl": "https://openagent3.xyz/downloads/cursor-ide-agent",
    "agentUrl": "https://openagent3.xyz/skills/cursor-ide-agent/agent",
    "manifestUrl": "https://openagent3.xyz/skills/cursor-ide-agent/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/cursor-ide-agent/agent.md"
  }
}