{ "schemaVersion": "1.0", "item": { "slug": "openclaw-skill-lazy-loader", "name": "OpenClaw Skill Lazy Loader", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader", "canonicalUrl": "https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/openclaw-skill-lazy-loader", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=openclaw-skill-lazy-loader", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "SKILLS.md.template", "AGENTS.md.template", "context_optimizer.py", "README.md", "SECURITY.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. 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/openclaw-skill-lazy-loader" }, "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/openclaw-skill-lazy-loader", "agentPageUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/agent", "manifestUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/agent.json", "briefUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/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": "OpenClaw Skill Lazy Loader", "body": "Stop loading every skill file at session start. Load what you need, when you need it — and cut your token usage by 40–70%." }, { "title": "The Problem", "body": "Most OpenClaw agents load their entire skill library at startup:\n\n# AGENTS.md (naive approach)\nRead ALL of these before starting:\n- skills/python/SKILL.md\n- skills/git/SKILL.md\n- skills/docker/SKILL.md\n- skills/aws/SKILL.md\n- skills/browser/SKILL.md\n... (20 more)\n\nEach session burns 3,000–15,000 tokens just loading context that may never be used. At scale, this is your biggest cost." }, { "title": "The Solution: Lazy Loading", "body": "Instead of loading skills upfront, agents check a SKILLS catalog (a lightweight index) and load individual skill files only when a task requires them.\n\nBefore: Load 20 skill files = ~12,000 tokens/session\nAfter: Load catalog (300 tokens) + 1–2 relevant skills (~800 tokens) = ~1,100 tokens/session\n\nThat's an 89% reduction on context loading alone." }, { "title": "Step 1: Create Your SKILLS Catalog", "body": "Create SKILLS.md in your agent workspace — a lightweight index of all available skills:\n\n# Available Skills\n\n| Skill | File | Use When |\n|-------|------|----------|\n| Python | skills/python/SKILL.md | Writing/debugging Python code |\n| Git | skills/git/SKILL.md | Git operations, PRs, branches |\n| Docker | skills/docker/SKILL.md | Containers, images, compose |\n| Browser | skills/browser/SKILL.md | Web scraping, UI automation |\n| AWS | skills/aws/SKILL.md | Cloud deployments, S3, Lambda |\n\nThis catalog is the ONLY file loaded at session start. ~200–400 tokens instead of 10,000+.\n\nSee SKILLS.md.template for a complete starter template." }, { "title": "Step 2: Update Your AGENTS.md", "body": "Replace bulk loading with the catalog pattern:\n\n## Skills\n\nAt session start: Read SKILLS.md (the index only).\nWhen a task needs a skill: Read the specific SKILL.md for that skill.\nNever load all skills upfront.\n\n### Loading Decision\nBefore loading any skill file:\n1. Does the current task need it? (yes → load it, no → skip)\n2. Has it already been loaded this session? (yes → skip, no → load once)\n\nSee AGENTS.md.template for the full recommended AGENTS.md skills section." }, { "title": "Step 3: Use the Context Optimizer (Optional)", "body": "The included context_optimizer.py analyzes your task description and recommends which skills to load:\n\npython3 context_optimizer.py recommend \"Write a Python script to push to S3\"\n# Output:\n# Recommended skills to load:\n# - skills/python/SKILL.md (confidence: high — Python task)\n# - skills/aws/SKILL.md (confidence: high — S3 mentioned)\n# - skills/git/SKILL.md (confidence: low — skip unless pushing to GitHub)" }, { "title": "Step 4: Apply to Memory Files Too", "body": "The same pattern works for memory and context files:\n\n## Memory Loading (AGENTS.md)\n\nAt session start: Read MEMORY.md (summary only).\nLoad daily files (memory/YYYY-MM-DD.md) only when:\n- User asks about past work\n- Task references a specific date or project\n- Debugging requires historical context" }, { "title": "Token Savings Calculator", "body": "ScenarioBeforeAfterSavings5 skills loaded~3,000 tokens~600 tokens80%10 skills loaded~6,500 tokens~750 tokens88%20 skills loaded~13,000 tokens~900 tokens93%+Memory files (5)+4,000 tokens+400 tokens90%\n\nEstimates based on average SKILL.md size of ~600 tokens. Catalog averages ~150 tokens." }, { "title": "Integration with Token Optimizer", "body": "This skill pairs directly with OpenClaw Token Optimizer. Lazy loading handles context loading costs; Token Optimizer handles model routing, heartbeat budgeting, and runtime costs. Together they cover the full token lifecycle.\n\nInstall both:\n\nclawhub install openclaw-skill-lazy-loader\nclawhub install openclaw-token-optimizer" }, { "title": "Files in This Skill", "body": "FilePurposeSKILL.mdThis guideSKILLS.md.templateStarter SKILLS catalog templateAGENTS.md.templateLazy loading AGENTS.md sectioncontext_optimizer.pyCLI helper — recommends skills to load per taskREADME.mdClawHub listing descriptionSECURITY.mdSecurity audit and script disclosure.clawhubsafeFile integrity manifest" }, { "title": "Quick Start (5 minutes)", "body": "# 1. Install\nclawhub install openclaw-skill-lazy-loader\n\n# 2. Copy templates to your agent workspace\ncp ~/.openclaw/skills/openclaw-skill-lazy-loader/SKILLS.md.template ~/my-agent/SKILLS.md\ncp ~/.openclaw/skills/openclaw-skill-lazy-loader/AGENTS.md.template ~/my-agent/AGENTS.lazy.md\n\n# 3. Edit SKILLS.md — fill in your actual skills\n# 4. Merge AGENTS.lazy.md into your AGENTS.md\n# 5. Test with context_optimizer.py\npython3 ~/.openclaw/skills/openclaw-skill-lazy-loader/context_optimizer.py recommend \"your next task\"\n\nBy M Asif Rahman (@Asif2BD) · Apache 2.0 · https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader" } ], "body": "OpenClaw Skill Lazy Loader\n\nStop loading every skill file at session start. Load what you need, when you need it — and cut your token usage by 40–70%.\n\nThe Problem\n\nMost OpenClaw agents load their entire skill library at startup:\n\n# AGENTS.md (naive approach)\nRead ALL of these before starting:\n- skills/python/SKILL.md\n- skills/git/SKILL.md\n- skills/docker/SKILL.md\n- skills/aws/SKILL.md\n- skills/browser/SKILL.md\n... (20 more)\n\n\nEach session burns 3,000–15,000 tokens just loading context that may never be used. At scale, this is your biggest cost.\n\nThe Solution: Lazy Loading\n\nInstead of loading skills upfront, agents check a SKILLS catalog (a lightweight index) and load individual skill files only when a task requires them.\n\nBefore: Load 20 skill files = ~12,000 tokens/session After: Load catalog (300 tokens) + 1–2 relevant skills (~800 tokens) = ~1,100 tokens/session\n\nThat's an 89% reduction on context loading alone.\n\nImplementation Guide\nStep 1: Create Your SKILLS Catalog\n\nCreate SKILLS.md in your agent workspace — a lightweight index of all available skills:\n\n# Available Skills\n\n| Skill | File | Use When |\n|-------|------|----------|\n| Python | skills/python/SKILL.md | Writing/debugging Python code |\n| Git | skills/git/SKILL.md | Git operations, PRs, branches |\n| Docker | skills/docker/SKILL.md | Containers, images, compose |\n| Browser | skills/browser/SKILL.md | Web scraping, UI automation |\n| AWS | skills/aws/SKILL.md | Cloud deployments, S3, Lambda |\n\n\nThis catalog is the ONLY file loaded at session start. ~200–400 tokens instead of 10,000+.\n\nSee SKILLS.md.template for a complete starter template.\n\nStep 2: Update Your AGENTS.md\n\nReplace bulk loading with the catalog pattern:\n\n## Skills\n\nAt session start: Read SKILLS.md (the index only).\nWhen a task needs a skill: Read the specific SKILL.md for that skill.\nNever load all skills upfront.\n\n### Loading Decision\nBefore loading any skill file:\n1. Does the current task need it? (yes → load it, no → skip)\n2. Has it already been loaded this session? (yes → skip, no → load once)\n\n\nSee AGENTS.md.template for the full recommended AGENTS.md skills section.\n\nStep 3: Use the Context Optimizer (Optional)\n\nThe included context_optimizer.py analyzes your task description and recommends which skills to load:\n\npython3 context_optimizer.py recommend \"Write a Python script to push to S3\"\n# Output:\n# Recommended skills to load:\n# - skills/python/SKILL.md (confidence: high — Python task)\n# - skills/aws/SKILL.md (confidence: high — S3 mentioned)\n# - skills/git/SKILL.md (confidence: low — skip unless pushing to GitHub)\n\nStep 4: Apply to Memory Files Too\n\nThe same pattern works for memory and context files:\n\n## Memory Loading (AGENTS.md)\n\nAt session start: Read MEMORY.md (summary only).\nLoad daily files (memory/YYYY-MM-DD.md) only when:\n- User asks about past work\n- Task references a specific date or project\n- Debugging requires historical context\n\nToken Savings Calculator\nScenario\tBefore\tAfter\tSavings\n5 skills loaded\t~3,000 tokens\t~600 tokens\t80%\n10 skills loaded\t~6,500 tokens\t~750 tokens\t88%\n20 skills loaded\t~13,000 tokens\t~900 tokens\t93%\n+Memory files (5)\t+4,000 tokens\t+400 tokens\t90%\n\nEstimates based on average SKILL.md size of ~600 tokens. Catalog averages ~150 tokens.\n\nIntegration with Token Optimizer\n\nThis skill pairs directly with OpenClaw Token Optimizer. Lazy loading handles context loading costs; Token Optimizer handles model routing, heartbeat budgeting, and runtime costs. Together they cover the full token lifecycle.\n\nInstall both:\n\nclawhub install openclaw-skill-lazy-loader\nclawhub install openclaw-token-optimizer\n\nFiles in This Skill\nFile\tPurpose\nSKILL.md\tThis guide\nSKILLS.md.template\tStarter SKILLS catalog template\nAGENTS.md.template\tLazy loading AGENTS.md section\ncontext_optimizer.py\tCLI helper — recommends skills to load per task\nREADME.md\tClawHub listing description\nSECURITY.md\tSecurity audit and script disclosure\n.clawhubsafe\tFile integrity manifest\nQuick Start (5 minutes)\n# 1. Install\nclawhub install openclaw-skill-lazy-loader\n\n# 2. Copy templates to your agent workspace\ncp ~/.openclaw/skills/openclaw-skill-lazy-loader/SKILLS.md.template ~/my-agent/SKILLS.md\ncp ~/.openclaw/skills/openclaw-skill-lazy-loader/AGENTS.md.template ~/my-agent/AGENTS.lazy.md\n\n# 3. Edit SKILLS.md — fill in your actual skills\n# 4. Merge AGENTS.lazy.md into your AGENTS.md\n# 5. Test with context_optimizer.py\npython3 ~/.openclaw/skills/openclaw-skill-lazy-loader/context_optimizer.py recommend \"your next task\"\n\n\nBy M Asif Rahman (@Asif2BD) · Apache 2.0 · https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader", "publisherUrl": "https://clawhub.ai/Asif2BD/openclaw-skill-lazy-loader", "owner": "Asif2BD", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader", "downloadUrl": "https://openagent3.xyz/downloads/openclaw-skill-lazy-loader", "agentUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/agent", "manifestUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/agent.json", "briefUrl": "https://openagent3.xyz/skills/openclaw-skill-lazy-loader/agent.md" } }