{
  "schemaVersion": "1.0",
  "item": {
    "slug": "octoflow",
    "name": "Octoflow",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/mikedconcepcion/octoflow",
    "canonicalUrl": "https://clawhub.ai/mikedconcepcion/octoflow",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/octoflow",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=octoflow",
    "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/octoflow"
    },
    "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/octoflow",
    "agentPageUrl": "https://openagent3.xyz/skills/octoflow/agent",
    "manifestUrl": "https://openagent3.xyz/skills/octoflow/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/octoflow/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": "OctoFlow",
        "body": "GPU-native programming language. Describe tasks in English, run them on any GPU via Vulkan."
      },
      {
        "title": "When to Use This Skill",
        "body": "Use this skill when the user says:\n\n\"sort a million numbers on GPU\" / \"benchmark GPU performance\"\n\"load this CSV and show me statistics\" / \"analyze my data\"\n\"cluster my dataset\" / \"run K-means\" / \"train a classifier\"\n\"blur this image\" / \"resize this BMP\" / \"encode a GIF\"\n\"plot height vs weight\" / \"create a scatter plot\"\n\"calculate the Sharpe ratio\" / \"compute correlation\"\n\"find primes\" / \"generate random numbers on GPU\"\n\"run linear regression on my dataset\"\n\"run a LLM on my GPU\" / \"inference on my GPU\"\n\nDo NOT use this skill when:\n\nThe user wants Python/JavaScript/Rust code (use the appropriate language tool)\nThe task doesn't benefit from GPU acceleration or OctoFlow's built-in functions\nThe user explicitly asks for a different language"
      },
      {
        "title": "Chat mode (natural language to running code)",
        "body": "octoflow chat \"sort 1M numbers on GPU\""
      },
      {
        "title": "Run a .flow script",
        "body": "octoflow run program.flow"
      },
      {
        "title": "Run with permissions (sandboxed by default)",
        "body": "# Allow reading data files\noctoflow run analysis.flow --allow-read=./data\n\n# Allow network access to specific domain\noctoflow chat \"fetch weather data\" --allow-net=api.weather.com\n\n# Allow writing output files\noctoflow run report.flow --allow-read=./data --allow-write=./output"
      },
      {
        "title": "Security Model",
        "body": "OctoFlow uses Deno-style permissions. Everything is denied by default.\n\nPermissionDefaultHow to enableExampleFile readDENIED--allow-read=./dataRead CSV from ./data onlyFile writeDENIED--allow-write=./outputWrite results to ./output onlyNetworkDENIED--allow-net=api.example.comFetch from one domain onlyProcess execDENIED--allow-exec=curlAllow curl only\n\nWithout flags, OctoFlow can only read .flow source files and print to stdout.\nNo file access, no network, no subprocesses unless the user explicitly opts in."
      },
      {
        "title": "MCP Server",
        "body": "OctoFlow can run as an MCP server for AI agent integration:\n\noctoflow mcp-serve\n\nAdd to your OpenClaw, Claude Desktop, or Cursor config:\n\n{\"mcpServers\": {\"octoflow\": {\"command\": \"octoflow\", \"args\": [\"mcp-serve\"]}}}"
      },
      {
        "title": "Available Tools",
        "body": "ToolDescriptionoctoflow_runExecute OctoFlow code directlyoctoflow_chatNatural language to GPU codeoctoflow_checkValidate .flow syntaxoctoflow_gpu_sortGPU-accelerated sortingoctoflow_gpu_statsGPU statistical operationsoctoflow_imageImage processing (BMP, GIF)octoflow_csvCSV data analysis"
      },
      {
        "title": "Data Analysis",
        "body": "# User: \"analyze sales.csv and show trends\"\noctoflow chat \"load sales.csv, compute monthly averages, and plot the trend\" --allow-read=."
      },
      {
        "title": "GPU Compute",
        "body": "# User: \"sort a large dataset on GPU\"\noctoflow chat \"generate 1M random numbers on GPU and sort them\""
      },
      {
        "title": "Machine Learning",
        "body": "# User: \"cluster my customers\"\noctoflow chat \"load customers.csv, run K-means with 5 clusters, print cluster sizes\" --allow-read=."
      },
      {
        "title": "Image Processing",
        "body": "# User: \"blur this photo\"\noctoflow chat \"load photo.bmp, apply gaussian blur, save as blurred.bmp\" --allow-read=. --allow-write=."
      },
      {
        "title": "Statistics",
        "body": "# User: \"what's the correlation between these columns?\"\noctoflow chat \"load data.csv, compute Pearson correlation between col1 and col2\" --allow-read=."
      },
      {
        "title": "Key Capabilities",
        "body": "FeatureDetailBuiltins210+ built-in functionsStdlib445 modules across 28 domainsGPU kernels150 Vulkan compute shadersGPU VMLoom Engine — 5 SSBOs, indirect dispatch, layer streamingGPU supportAny Vulkan GPU (NVIDIA, AMD, Intel)Binary size4.5 MB, zero dependenciesChat modeEnglish to code with auto-fix loop (max 3 retries)Errors69 structured error codes with auto-fix suggestionsMCP Server7 structured tools via JSON-RPC 2.0PlatformsWindows, Linux, macOS (Apple Silicon)"
      },
      {
        "title": "Data Storage",
        "body": "OctoFlow optionally saves your preferences to ~/.octoflow/ (user-level) and .octoflow/ (per-project).\n\nContents: which stdlib modules you use frequently and corrections from previous sessions.\n\nNo telemetry. No data is sent anywhere.\nNo network calls unless you explicitly use --allow-net.\nAll data stays local on your machine.\nDisable entirely with --no-memory flag — nothing is saved.\nProject config via OCTOFLOW.md in your project root (like .eslintrc or pyproject.toml)."
      },
      {
        "title": "Download (recommended)",
        "body": "PlatformFileSHA-256Windows x64octoflow-v1.5.8-x86_64-windows.zip2b26049565a2bfd2b1c4a1c103f2a64cd864dd14da619bd7be750ad3c6b356f2Linux x64octoflow-v1.5.8-x86_64-linux.tar.gzd7306fc1f5a9a733a66ae3a4d5f3b145670efa7a079302935d867b4b75551845macOS (Apple Silicon)octoflow-v1.5.8-aarch64-macos.tar.gz33808c330dc5f08eb0008b52ecfb5f0ea532fb71b1c6996075c09b33dc5d8fd2\n\nVerify: sha256sum octoflow-v1.5.8-* (full checksums in SHA256SUMS.txt).\n\nUnzip/extract, add to PATH. No installer needed."
      },
      {
        "title": "Links",
        "body": "GitHub\nDocumentation\nReleases"
      }
    ],
    "body": "OctoFlow\n\nGPU-native programming language. Describe tasks in English, run them on any GPU via Vulkan.\n\nWhen to Use This Skill\n\nUse this skill when the user says:\n\n\"sort a million numbers on GPU\" / \"benchmark GPU performance\"\n\"load this CSV and show me statistics\" / \"analyze my data\"\n\"cluster my dataset\" / \"run K-means\" / \"train a classifier\"\n\"blur this image\" / \"resize this BMP\" / \"encode a GIF\"\n\"plot height vs weight\" / \"create a scatter plot\"\n\"calculate the Sharpe ratio\" / \"compute correlation\"\n\"find primes\" / \"generate random numbers on GPU\"\n\"run linear regression on my dataset\"\n\"run a LLM on my GPU\" / \"inference on my GPU\"\n\nDo NOT use this skill when:\n\nThe user wants Python/JavaScript/Rust code (use the appropriate language tool)\nThe task doesn't benefit from GPU acceleration or OctoFlow's built-in functions\nThe user explicitly asks for a different language\nHow to Run OctoFlow\nChat mode (natural language to running code)\noctoflow chat \"sort 1M numbers on GPU\"\n\nRun a .flow script\noctoflow run program.flow\n\nRun with permissions (sandboxed by default)\n# Allow reading data files\noctoflow run analysis.flow --allow-read=./data\n\n# Allow network access to specific domain\noctoflow chat \"fetch weather data\" --allow-net=api.weather.com\n\n# Allow writing output files\noctoflow run report.flow --allow-read=./data --allow-write=./output\n\nSecurity Model\n\nOctoFlow uses Deno-style permissions. Everything is denied by default.\n\nPermission\tDefault\tHow to enable\tExample\nFile read\tDENIED\t--allow-read=./data\tRead CSV from ./data only\nFile write\tDENIED\t--allow-write=./output\tWrite results to ./output only\nNetwork\tDENIED\t--allow-net=api.example.com\tFetch from one domain only\nProcess exec\tDENIED\t--allow-exec=curl\tAllow curl only\n\nWithout flags, OctoFlow can only read .flow source files and print to stdout. No file access, no network, no subprocesses unless the user explicitly opts in.\n\nMCP Server\n\nOctoFlow can run as an MCP server for AI agent integration:\n\noctoflow mcp-serve\n\n\nAdd to your OpenClaw, Claude Desktop, or Cursor config:\n\n{\"mcpServers\": {\"octoflow\": {\"command\": \"octoflow\", \"args\": [\"mcp-serve\"]}}}\n\nAvailable Tools\nTool\tDescription\noctoflow_run\tExecute OctoFlow code directly\noctoflow_chat\tNatural language to GPU code\noctoflow_check\tValidate .flow syntax\noctoflow_gpu_sort\tGPU-accelerated sorting\noctoflow_gpu_stats\tGPU statistical operations\noctoflow_image\tImage processing (BMP, GIF)\noctoflow_csv\tCSV data analysis\nCommon Patterns\nData Analysis\n# User: \"analyze sales.csv and show trends\"\noctoflow chat \"load sales.csv, compute monthly averages, and plot the trend\" --allow-read=.\n\nGPU Compute\n# User: \"sort a large dataset on GPU\"\noctoflow chat \"generate 1M random numbers on GPU and sort them\"\n\nMachine Learning\n# User: \"cluster my customers\"\noctoflow chat \"load customers.csv, run K-means with 5 clusters, print cluster sizes\" --allow-read=.\n\nImage Processing\n# User: \"blur this photo\"\noctoflow chat \"load photo.bmp, apply gaussian blur, save as blurred.bmp\" --allow-read=. --allow-write=.\n\nStatistics\n# User: \"what's the correlation between these columns?\"\noctoflow chat \"load data.csv, compute Pearson correlation between col1 and col2\" --allow-read=.\n\nKey Capabilities\nFeature\tDetail\nBuiltins\t210+ built-in functions\nStdlib\t445 modules across 28 domains\nGPU kernels\t150 Vulkan compute shaders\nGPU VM\tLoom Engine — 5 SSBOs, indirect dispatch, layer streaming\nGPU support\tAny Vulkan GPU (NVIDIA, AMD, Intel)\nBinary size\t4.5 MB, zero dependencies\nChat mode\tEnglish to code with auto-fix loop (max 3 retries)\nErrors\t69 structured error codes with auto-fix suggestions\nMCP Server\t7 structured tools via JSON-RPC 2.0\nPlatforms\tWindows, Linux, macOS (Apple Silicon)\nData Storage\n\nOctoFlow optionally saves your preferences to ~/.octoflow/ (user-level) and .octoflow/ (per-project).\n\nContents: which stdlib modules you use frequently and corrections from previous sessions.\n\nNo telemetry. No data is sent anywhere.\nNo network calls unless you explicitly use --allow-net.\nAll data stays local on your machine.\nDisable entirely with --no-memory flag — nothing is saved.\nProject config via OCTOFLOW.md in your project root (like .eslintrc or pyproject.toml).\nInstall\nDownload (recommended)\nPlatform\tFile\tSHA-256\nWindows x64\toctoflow-v1.5.8-x86_64-windows.zip\t2b26049565a2bfd2b1c4a1c103f2a64cd864dd14da619bd7be750ad3c6b356f2\nLinux x64\toctoflow-v1.5.8-x86_64-linux.tar.gz\td7306fc1f5a9a733a66ae3a4d5f3b145670efa7a079302935d867b4b75551845\nmacOS (Apple Silicon)\toctoflow-v1.5.8-aarch64-macos.tar.gz\t33808c330dc5f08eb0008b52ecfb5f0ea532fb71b1c6996075c09b33dc5d8fd2\n\nVerify: sha256sum octoflow-v1.5.8-* (full checksums in SHA256SUMS.txt).\n\nUnzip/extract, add to PATH. No installer needed.\n\nLinks\nGitHub\nDocumentation\nReleases"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/mikedconcepcion/octoflow",
    "publisherUrl": "https://clawhub.ai/mikedconcepcion/octoflow",
    "owner": "mikedconcepcion",
    "version": "1.5.8",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/octoflow",
    "downloadUrl": "https://openagent3.xyz/downloads/octoflow",
    "agentUrl": "https://openagent3.xyz/skills/octoflow/agent",
    "manifestUrl": "https://openagent3.xyz/skills/octoflow/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/octoflow/agent.md"
  }
}