{
  "schemaVersion": "1.0",
  "item": {
    "slug": "wandb-monitor",
    "name": "Weights & Biases Monitor",
    "source": "tencent",
    "type": "skill",
    "category": "数据分析",
    "sourceUrl": "https://clawhub.ai/chrisvoncsefalvay/wandb-monitor",
    "canonicalUrl": "https://clawhub.ai/chrisvoncsefalvay/wandb-monitor",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/wandb-monitor",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=wandb-monitor",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "scripts/characterize_run.py",
      "scripts/check_runs.py",
      "scripts/compare_runs.py",
      "scripts/run_details.py",
      "scripts/watch_runs.py"
    ],
    "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": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.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/wandb-monitor"
    },
    "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/wandb-monitor",
    "agentPageUrl": "https://openagent3.xyz/skills/wandb-monitor/agent",
    "manifestUrl": "https://openagent3.xyz/skills/wandb-monitor/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/wandb-monitor/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": "Weights & Biases",
        "body": "Monitor, analyze, and compare W&B training runs."
      },
      {
        "title": "Setup",
        "body": "wandb login\n# Or set WANDB_API_KEY in environment"
      },
      {
        "title": "Characterize a Run (Full Health Analysis)",
        "body": "~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/characterize_run.py ENTITY/PROJECT/RUN_ID\n\nAnalyzes:\n\nLoss curve trend (start → current, % change, direction)\nGradient norm health (exploding/vanishing detection)\nEval metrics (if present)\nStall detection (heartbeat age)\nProgress & ETA estimate\nConfig highlights\nOverall health verdict\n\nOptions: --json for machine-readable output."
      },
      {
        "title": "Watch All Running Jobs",
        "body": "~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/watch_runs.py ENTITY [--projects p1,p2]\n\nQuick health summary of all running jobs plus recent failures/completions. Ideal for morning briefings.\n\nOptions:\n\n--projects p1,p2 — Specific projects to check\n--all-projects — Check all projects\n--hours N — Hours to look back for finished runs (default: 24)\n--json — Machine-readable output"
      },
      {
        "title": "Compare Two Runs",
        "body": "~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/compare_runs.py ENTITY/PROJECT/RUN_A ENTITY/PROJECT/RUN_B\n\nSide-by-side comparison:\n\nConfig differences (highlights important params)\nLoss curves at same steps\nGradient norm comparison\nEval metrics\nPerformance (tokens/sec, steps/hour)\nWinner verdict"
      },
      {
        "title": "Python API Quick Reference",
        "body": "import wandb\napi = wandb.Api()\n\n# Get runs\nruns = api.runs(\"entity/project\", {\"state\": \"running\"})\n\n# Run properties\nrun.state      # running | finished | failed | crashed | canceled\nrun.name       # display name\nrun.id         # unique identifier\nrun.summary    # final/current metrics\nrun.config     # hyperparameters\nrun.heartbeat_at # stall detection\n\n# Get history\nhistory = list(run.scan_history(keys=[\"train/loss\", \"train/grad_norm\"]))"
      },
      {
        "title": "Metric Key Variations",
        "body": "Scripts handle these automatically:\n\nLoss: train/loss, loss, train_loss, training_loss\nGradients: train/grad_norm, grad_norm, gradient_norm\nSteps: train/global_step, global_step, step, _step\nEval: eval/loss, eval_loss, eval/accuracy, eval_acc"
      },
      {
        "title": "Health Thresholds",
        "body": "Gradients > 10: Exploding (critical)\nGradients > 5: Spiky (warning)\nGradients < 0.0001: Vanishing (warning)\nHeartbeat > 30min: Stalled (critical)\nHeartbeat > 10min: Slow (warning)"
      },
      {
        "title": "Integration Notes",
        "body": "For morning briefings, use watch_runs.py --json and parse the output.\n\nFor detailed analysis of a specific run, use characterize_run.py.\n\nFor A/B testing or hyperparameter comparisons, use compare_runs.py."
      }
    ],
    "body": "Weights & Biases\n\nMonitor, analyze, and compare W&B training runs.\n\nSetup\nwandb login\n# Or set WANDB_API_KEY in environment\n\nScripts\nCharacterize a Run (Full Health Analysis)\n~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/characterize_run.py ENTITY/PROJECT/RUN_ID\n\n\nAnalyzes:\n\nLoss curve trend (start → current, % change, direction)\nGradient norm health (exploding/vanishing detection)\nEval metrics (if present)\nStall detection (heartbeat age)\nProgress & ETA estimate\nConfig highlights\nOverall health verdict\n\nOptions: --json for machine-readable output.\n\nWatch All Running Jobs\n~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/watch_runs.py ENTITY [--projects p1,p2]\n\n\nQuick health summary of all running jobs plus recent failures/completions. Ideal for morning briefings.\n\nOptions:\n\n--projects p1,p2 — Specific projects to check\n--all-projects — Check all projects\n--hours N — Hours to look back for finished runs (default: 24)\n--json — Machine-readable output\nCompare Two Runs\n~/clawd/venv/bin/python3 ~/clawd/skills/wandb/scripts/compare_runs.py ENTITY/PROJECT/RUN_A ENTITY/PROJECT/RUN_B\n\n\nSide-by-side comparison:\n\nConfig differences (highlights important params)\nLoss curves at same steps\nGradient norm comparison\nEval metrics\nPerformance (tokens/sec, steps/hour)\nWinner verdict\nPython API Quick Reference\nimport wandb\napi = wandb.Api()\n\n# Get runs\nruns = api.runs(\"entity/project\", {\"state\": \"running\"})\n\n# Run properties\nrun.state      # running | finished | failed | crashed | canceled\nrun.name       # display name\nrun.id         # unique identifier\nrun.summary    # final/current metrics\nrun.config     # hyperparameters\nrun.heartbeat_at # stall detection\n\n# Get history\nhistory = list(run.scan_history(keys=[\"train/loss\", \"train/grad_norm\"]))\n\nMetric Key Variations\n\nScripts handle these automatically:\n\nLoss: train/loss, loss, train_loss, training_loss\nGradients: train/grad_norm, grad_norm, gradient_norm\nSteps: train/global_step, global_step, step, _step\nEval: eval/loss, eval_loss, eval/accuracy, eval_acc\nHealth Thresholds\nGradients > 10: Exploding (critical)\nGradients > 5: Spiky (warning)\nGradients < 0.0001: Vanishing (warning)\nHeartbeat > 30min: Stalled (critical)\nHeartbeat > 10min: Slow (warning)\nIntegration Notes\n\nFor morning briefings, use watch_runs.py --json and parse the output.\n\nFor detailed analysis of a specific run, use characterize_run.py.\n\nFor A/B testing or hyperparameter comparisons, use compare_runs.py."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/chrisvoncsefalvay/wandb-monitor",
    "publisherUrl": "https://clawhub.ai/chrisvoncsefalvay/wandb-monitor",
    "owner": "chrisvoncsefalvay",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/wandb-monitor",
    "downloadUrl": "https://openagent3.xyz/downloads/wandb-monitor",
    "agentUrl": "https://openagent3.xyz/skills/wandb-monitor/agent",
    "manifestUrl": "https://openagent3.xyz/skills/wandb-monitor/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/wandb-monitor/agent.md"
  }
}