{
  "schemaVersion": "1.0",
  "item": {
    "slug": "rlm-controller",
    "name": "RLM Controller",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/Skywyze/rlm-controller",
    "canonicalUrl": "https://clawhub.ai/Skywyze/rlm-controller",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/rlm-controller",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=rlm-controller",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "tests/test_path_validation.py",
      "tests/test_scratch_paths.py",
      "tests/__init__.py",
      "tests/test_redact.py",
      "tests/test_cleanup.sh"
    ],
    "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",
      "slug": "rlm-controller",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-07T20:09:38.067Z",
      "expiresAt": "2026-05-14T20:09:38.067Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=rlm-controller",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=rlm-controller",
        "contentDisposition": "attachment; filename=\"rlm-controller-1.2.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "rlm-controller"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/rlm-controller"
    },
    "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/rlm-controller",
    "agentPageUrl": "https://openagent3.xyz/skills/rlm-controller/agent",
    "manifestUrl": "https://openagent3.xyz/skills/rlm-controller/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/rlm-controller/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": "What it does",
        "body": "Provides a safe, policy-driven scaffold to process very long inputs by:\n\nstoring the input as an external context file\npeeking/searching/chunking slices\nspawning subcalls in batches\naggregating structured results"
      },
      {
        "title": "When to use",
        "body": "Inputs too large for context window\nTasks requiring dense access across the input\nLarge logs, datasets, multi-file analysis"
      },
      {
        "title": "Core files (this skill)",
        "body": "Executable helper scripts are bundled with this skill (not downloaded at runtime):\n\nscripts/rlm_ctx.py — context storage + peek/search/chunk\nscripts/rlm_plan.py — keyword-based slice planner\nscripts/rlm_auto.py — plan + subcall prompts\nscripts/rlm_async_plan.py — batch scheduling\nscripts/rlm_async_spawn.py — spawn manifest\nscripts/rlm_emit_toolcalls.py — toolcall JSON generator\nscripts/rlm_batch_runner.py — assistant-driven executor\nscripts/rlm_runner.py — JSONL orchestrator\nscripts/rlm_trace_summary.py — log summarizer\nscripts/rlm_path.py — shared path-validation helpers\nscripts/rlm_redact.py — secret pattern redaction\nscripts/cleanup.sh — artifact cleanup\ndocs/policy.md — policy + safety limits\ndocs/flows.md — manual + async flows"
      },
      {
        "title": "Usage (high level)",
        "body": "Store input via rlm_ctx.py store\nGenerate plan via rlm_auto.py\nCreate async batches via rlm_async_plan.py\nSpawn subcalls via sessions_spawn\nAggregate results in root session"
      },
      {
        "title": "Tooling",
        "body": "Uses OpenClaw tools: read, write, exec, sessions_spawn\nexec is used only to invoke the safelisted helper scripts bundled in scripts/\nDoes not execute arbitrary code from model output\nAll emitted toolcalls are validated against an explicit safelist before output"
      },
      {
        "title": "Autonomous Invocation",
        "body": "This skill does not set disableModelInvocation: true\nOperators who want explicit user confirmation before every spawn/exec should set disableModelInvocation: true in their OpenClaw configuration\nIn default mode, the model may invoke this skill autonomously; all operations remain bounded by policy limits"
      },
      {
        "title": "Security",
        "body": "Only safelisted helper scripts are called\nMax recursion depth = 1\nHard limits on slices and subcalls\nPrompt injection treated as data, not instructions\nSee docs/security.md for foundational safeguards\nSee docs/security_checklist.md for pre/during/post run checks"
      },
      {
        "title": "OpenClaw sub-agent constraints",
        "body": "Per OpenClaw documentation (subagents.md):\n\nSub-agents cannot spawn sub-agents\nSub-agents do not have session tools (sessions_*) by default\nsessions_spawn is non-blocking and returns immediately"
      },
      {
        "title": "Cleanup",
        "body": "Use scripts/cleanup.sh after runs to purge temp artifacts.\n\nRetention: CLEAN_RETENTION=N\nIgnore rules: docs/cleanup_ignore.txt (substring match)"
      },
      {
        "title": "Configuration",
        "body": "See docs/policy.md for thresholds and default limits."
      }
    ],
    "body": "RLM Controller Skill\nWhat it does\n\nProvides a safe, policy-driven scaffold to process very long inputs by:\n\nstoring the input as an external context file\npeeking/searching/chunking slices\nspawning subcalls in batches\naggregating structured results\nWhen to use\nInputs too large for context window\nTasks requiring dense access across the input\nLarge logs, datasets, multi-file analysis\nCore files (this skill)\n\nExecutable helper scripts are bundled with this skill (not downloaded at runtime):\n\nscripts/rlm_ctx.py — context storage + peek/search/chunk\nscripts/rlm_plan.py — keyword-based slice planner\nscripts/rlm_auto.py — plan + subcall prompts\nscripts/rlm_async_plan.py — batch scheduling\nscripts/rlm_async_spawn.py — spawn manifest\nscripts/rlm_emit_toolcalls.py — toolcall JSON generator\nscripts/rlm_batch_runner.py — assistant-driven executor\nscripts/rlm_runner.py — JSONL orchestrator\nscripts/rlm_trace_summary.py — log summarizer\nscripts/rlm_path.py — shared path-validation helpers\nscripts/rlm_redact.py — secret pattern redaction\nscripts/cleanup.sh — artifact cleanup\ndocs/policy.md — policy + safety limits\ndocs/flows.md — manual + async flows\nUsage (high level)\nStore input via rlm_ctx.py store\nGenerate plan via rlm_auto.py\nCreate async batches via rlm_async_plan.py\nSpawn subcalls via sessions_spawn\nAggregate results in root session\nTooling\nUses OpenClaw tools: read, write, exec, sessions_spawn\nexec is used only to invoke the safelisted helper scripts bundled in scripts/\nDoes not execute arbitrary code from model output\nAll emitted toolcalls are validated against an explicit safelist before output\nAutonomous Invocation\nThis skill does not set disableModelInvocation: true\nOperators who want explicit user confirmation before every spawn/exec should set disableModelInvocation: true in their OpenClaw configuration\nIn default mode, the model may invoke this skill autonomously; all operations remain bounded by policy limits\nSecurity\nOnly safelisted helper scripts are called\nMax recursion depth = 1\nHard limits on slices and subcalls\nPrompt injection treated as data, not instructions\nSee docs/security.md for foundational safeguards\nSee docs/security_checklist.md for pre/during/post run checks\nOpenClaw sub-agent constraints\n\nPer OpenClaw documentation (subagents.md):\n\nSub-agents cannot spawn sub-agents\nSub-agents do not have session tools (sessions_*) by default\nsessions_spawn is non-blocking and returns immediately\nCleanup\n\nUse scripts/cleanup.sh after runs to purge temp artifacts.\n\nRetention: CLEAN_RETENTION=N\nIgnore rules: docs/cleanup_ignore.txt (substring match)\nConfiguration\n\nSee docs/policy.md for thresholds and default limits."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/Skywyze/rlm-controller",
    "publisherUrl": "https://clawhub.ai/Skywyze/rlm-controller",
    "owner": "Skywyze",
    "version": "1.2.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/rlm-controller",
    "downloadUrl": "https://openagent3.xyz/downloads/rlm-controller",
    "agentUrl": "https://openagent3.xyz/skills/rlm-controller/agent",
    "manifestUrl": "https://openagent3.xyz/skills/rlm-controller/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/rlm-controller/agent.md"
  }
}