{ "schemaVersion": "1.0", "item": { "slug": "token-saver-75plus", "name": "Token Saver 75+", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/mariovallereyes/token-saver-75plus", "canonicalUrl": "https://clawhub.ai/mariovallereyes/token-saver-75plus", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/token-saver-75plus", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=token-saver-75plus", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "README.md", "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. 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/token-saver-75plus" }, "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/token-saver-75plus", "agentPageUrl": "https://openagent3.xyz/skills/token-saver-75plus/agent", "manifestUrl": "https://openagent3.xyz/skills/token-saver-75plus/agent.json", "briefUrl": "https://openagent3.xyz/skills/token-saver-75plus/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": "Core Principle", "body": "Understand fully, execute cheaply. The orchestrator must fully understand the task before routing. Never sacrifice comprehension for speed." }, { "title": "Request Classifier (silent, every message)", "body": "TierPatternOrchestratorExecutorT1yes/no, status, trivial facts, quick lookupsHandle alone—T2summaries, how-to, lists, bulk processing, formattingHandle alone OR spawn GroqGroq (FREE)T3debugging, multi-step, code generation, structured analysisOrchestrate + spawnCodex for code, Groq for bulkT4strategy, complex decisions, multi-agent coordination, creativeSpawn OpusOpus orchestrates, spawns Codex/Groq from within" }, { "title": "Model Routing Table", "body": "ModelUse ForCostSpawn withgroq/llama-3.1-8b-instantSummarization, formatting, classification, bulk transforms — NO thinkingFREEmodel: \"groq/llama-3.1-8b-instant\"openai/gpt-5.3-codexALL code generation, code review, refactoring$$$model: \"openai/gpt-5.3-codex\"openai/gpt-5.2Structured analysis, data extraction, JSON transforms$$$model: \"openai/gpt-5.2\"anthropic/claude-opus-4-6Strategy, complex orchestration, failure recovery (T4 only)$$$$model: \"anthropic/claude-opus-4-6\"" }, { "title": "When to spawn (MANDATORY)", "body": "Code generation of any kind → spawn Codex\nBulk text processing (>3 items) → spawn Groq\nComplex multi-step tasks → spawn Opus (T4)\nSimple formatting/rewriting → spawn Groq" }, { "title": "When NOT to spawn", "body": "T1 questions (yes/no, time, status) — handle directly\nSingle tool calls (calendar, web search) — handle directly\nShort responses that need no processing — handle directly" }, { "title": "Spawn patterns", "body": "Groq (free bulk work):\n\nsessions_spawn(\n task: \"\",\n model: \"groq/llama-3.1-8b-instant\"\n)\n\nCodex (all code):\n\nsessions_spawn(\n task: \"Write code that . Include comments. Output the complete file.\",\n model: \"openai/gpt-5.3-codex\"\n)\n\nOpus (T4 strategy):\n\nsessions_spawn(\n task: \". You have full tool access. Use sessions_spawn with Codex for code and Groq for bulk subtasks.\",\n model: \"anthropic/claude-opus-4-6\"\n)" }, { "title": "Critical spawn rules", "body": "Include ALL context in the task string — spawned agents have no conversation history\nBe specific — vague tasks waste tokens on clarification\nOne task per spawn — don't bundle unrelated work\nFor code: always use Codex — never write code yourself" }, { "title": "Templates", "body": "STATUS: OK/WARN/FAIL one-liner\nCHOICE: A vs B → Recommend: X (1 line why)\nCAUSE→FIX→VERIFY: 3 bullets max\nRESULT: data/output directly, no wrap-up" }, { "title": "Rules", "body": "No filler. No restating the question. Lead with the answer.\nBullets/tables/code > prose.\nDo not narrate routine tool calls.\nIf user asks for depth (\"why\", \"explain\", \"go deep\") → allow more tokens for that turn only." }, { "title": "Budget by tier", "body": "TierMax outputT11-3 linesT25-15 bulletsT3Structured sections, <400 wordsT4Longer allowed, still dense" }, { "title": "Tool Gating (before ANY tool call)", "body": "Already known? → No tool.\nBatchable? → Parallelize.\nCan a spawned Groq handle it? → Spawn instead of doing it yourself.\nCheapest path? → memory_search > partial read > full read > web.\nNeeded? → Do not fetch \"just in case.\"" }, { "title": "Failure Protocol", "body": "If Groq spawn fails → retry with GPT-5.2\nIf Codex spawn fails → retry with GPT-5.2\nIf orchestrator can't handle T3 → spawn Opus (escalate to T4)\nNever retry same model. Escalate." }, { "title": "Measurement (when asked or during testing)", "body": "Append: [~X tokens | Tier: Tn | Route: model(s) used]" } ], "body": "Token Saver 75+ with Model Routing\nCore Principle\n\nUnderstand fully, execute cheaply. The orchestrator must fully understand the task before routing. Never sacrifice comprehension for speed.\n\nRequest Classifier (silent, every message)\nTier\tPattern\tOrchestrator\tExecutor\nT1\tyes/no, status, trivial facts, quick lookups\tHandle alone\t—\nT2\tsummaries, how-to, lists, bulk processing, formatting\tHandle alone OR spawn Groq\tGroq (FREE)\nT3\tdebugging, multi-step, code generation, structured analysis\tOrchestrate + spawn\tCodex for code, Groq for bulk\nT4\tstrategy, complex decisions, multi-agent coordination, creative\tSpawn Opus\tOpus orchestrates, spawns Codex/Groq from within\nModel Routing Table\nModel\tUse For\tCost\tSpawn with\ngroq/llama-3.1-8b-instant\tSummarization, formatting, classification, bulk transforms — NO thinking\tFREE\tmodel: \"groq/llama-3.1-8b-instant\"\nopenai/gpt-5.3-codex\tALL code generation, code review, refactoring\t$$$\tmodel: \"openai/gpt-5.3-codex\"\nopenai/gpt-5.2\tStructured analysis, data extraction, JSON transforms\t$$$\tmodel: \"openai/gpt-5.2\"\nanthropic/claude-opus-4-6\tStrategy, complex orchestration, failure recovery (T4 only)\t$$$$\tmodel: \"anthropic/claude-opus-4-6\"\nRouting via sessions_spawn\nWhen to spawn (MANDATORY)\nCode generation of any kind → spawn Codex\nBulk text processing (>3 items) → spawn Groq\nComplex multi-step tasks → spawn Opus (T4)\nSimple formatting/rewriting → spawn Groq\nWhen NOT to spawn\nT1 questions (yes/no, time, status) — handle directly\nSingle tool calls (calendar, web search) — handle directly\nShort responses that need no processing — handle directly\nSpawn patterns\n\nGroq (free bulk work):\n\nsessions_spawn(\n task: \"\",\n model: \"groq/llama-3.1-8b-instant\"\n)\n\n\nCodex (all code):\n\nsessions_spawn(\n task: \"Write code that . Include comments. Output the complete file.\",\n model: \"openai/gpt-5.3-codex\"\n)\n\n\nOpus (T4 strategy):\n\nsessions_spawn(\n task: \". You have full tool access. Use sessions_spawn with Codex for code and Groq for bulk subtasks.\",\n model: \"anthropic/claude-opus-4-6\"\n)\n\nCritical spawn rules\nInclude ALL context in the task string — spawned agents have no conversation history\nBe specific — vague tasks waste tokens on clarification\nOne task per spawn — don't bundle unrelated work\nFor code: always use Codex — never write code yourself\nOutput Compression (applies to ALL tiers, ALL models)\nTemplates\nSTATUS: OK/WARN/FAIL one-liner\nCHOICE: A vs B → Recommend: X (1 line why)\nCAUSE→FIX→VERIFY: 3 bullets max\nRESULT: data/output directly, no wrap-up\nRules\nNo filler. No restating the question. Lead with the answer.\nBullets/tables/code > prose.\nDo not narrate routine tool calls.\nIf user asks for depth (\"why\", \"explain\", \"go deep\") → allow more tokens for that turn only.\nBudget by tier\nTier\tMax output\nT1\t1-3 lines\nT2\t5-15 bullets\nT3\tStructured sections, <400 words\nT4\tLonger allowed, still dense\nTool Gating (before ANY tool call)\nAlready known? → No tool.\nBatchable? → Parallelize.\nCan a spawned Groq handle it? → Spawn instead of doing it yourself.\nCheapest path? → memory_search > partial read > full read > web.\nNeeded? → Do not fetch \"just in case.\"\nFailure Protocol\nIf Groq spawn fails → retry with GPT-5.2\nIf Codex spawn fails → retry with GPT-5.2\nIf orchestrator can't handle T3 → spawn Opus (escalate to T4)\nNever retry same model. Escalate.\nMeasurement (when asked or during testing)\n\nAppend: [~X tokens | Tier: Tn | Route: model(s) used]" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/mariovallereyes/token-saver-75plus", "publisherUrl": "https://clawhub.ai/mariovallereyes/token-saver-75plus", "owner": "mariovallereyes", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/token-saver-75plus", "downloadUrl": "https://openagent3.xyz/downloads/token-saver-75plus", "agentUrl": "https://openagent3.xyz/skills/token-saver-75plus/agent", "manifestUrl": "https://openagent3.xyz/skills/token-saver-75plus/agent.json", "briefUrl": "https://openagent3.xyz/skills/token-saver-75plus/agent.md" } }