[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6JCTDVFAM4Q3DQ244EY43GA":3,"skill-same:01M6JCTDVFAM4Q3DQ244EY43GA":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":56,"install":57,"resources":70},"01M6JCTDVFAM4Q3DQ244EY43GA","cost-aware-llm-pipeline",null,"Cost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.","S3YED\u002Fappie-kit","S3YED","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F62743867?v=4","01M6JBKBRRVP6KMS0VWZ27T3R6","01M6JBKBX0KG30284PW1DBYJT1",14,"MIT","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fecc\u002Fcost-aware-llm-pipeline\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"pipeline","cost aware llm pipeline","cost","complexity","for","usage","retry","logic","caching","patterns","task","cost optimization","tracking","prompt","model","api","Budget","prompt caching","Optimization","routing","2026-11-10 10:12:30",false,"---\nname: cost-aware-llm-pipeline\ndescription: Cost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.\norigin: ECC\n---\n\n# Cost-Aware LLM Pipeline\n\nPatterns for controlling LLM API costs while maintaining quality. Combines model routing, budget tracking, retry logic, and prompt caching into a composable pipeline.\n\n## When to Activate\n\n- Building applications that call LLM APIs (Claude, GPT, etc.)\n- Processing batches of items with varying complexity\n- Need to stay within a budget for API spend\n- Optimizing cost without sacrificing quality on complex tasks\n\n## Core Concepts\n\n### 1. Model Routing by Task Complexity\n\nAutomatically select cheaper models for simple tasks, reserving expensive models for complex ones.\n\n```python\nMODEL_SONNET = \"claude-sonnet-4-6\"\nMODEL_HAIKU = \"claude-haiku-4-5-20251001\"\n\n_SONNET_TEXT_THRESHOLD = 10_000  # chars\n_SONNET_ITEM_THRESHOLD = 30     # items\n\ndef select_model(\n    text_length: int,\n    item_count: int,\n    force_model: str | None = None,\n) -> str:\n    \"\"\"Select model based on task complexity.\"\"\"\n    if force_model is not None:\n        return force_model\n    if text_length >= _SONNET_TEXT_THRESHOLD or item_count >= _SONNET_ITEM_THRESHOLD:\n        return MODEL_SONNET  # Complex task\n    return MODEL_HAIKU  # Simple task (3-4x cheaper)\n```\n\n### 2. Immutable Cost Tracking\n\nTrack cumulative spend with frozen dataclasses. Each API call returns a new tracker — never mutates state.\n\n```python\nfrom dataclasses import dataclass\n\n@dataclass(frozen=True, slots=True)\nclass CostRecord:\n    model: str\n    input_tokens: int\n    output_tokens: int\n    cost_usd: float\n\n@dataclass(frozen=True, slots=True)\nclass CostTracker:\n    budget_limit: float = 1.00\n    records: tuple[CostRecord, ...] = ()\n\n    def add(self, record: CostRecord) -> \"CostTracker\":\n        \"\"\"Return new tracker with added record (never mutates self).\"\"\"\n        return CostTracker(\n            budget_limit=self.budget_limit,\n            records=(*self.records, record),\n        )\n\n    @property\n    def total_cost(self) -> float:\n        return sum(r.cost_usd for r in self.records)\n\n    @property\n    def over_budget(self) -> bool:\n        return self.total_cost > self.budget_limit\n```\n\n### 3. Narrow Retry Logic\n\nRetry only on transient errors. Fail fast on authentication or bad request errors.\n\n```python\nfrom anthropic import (\n    APIConnectionError,\n    InternalServerError,\n    RateLimitError,\n)\n\n_RETRYABLE_ERRORS = (APIConnectionError, RateLimitError, InternalServerError)\n_MAX_RETRIES = 3\n\ndef call_with_retry(func, *, max_retries: int = _MAX_RETRIES):\n    \"\"\"Retry only on transient errors, fail fast on others.\"\"\"\n    for attempt in range(max_retries):\n        try:\n            return func()\n        except _RETRYABLE_ERRORS:\n            if attempt == max_retries - 1:\n                raise\n            time.sleep(2 ** attempt)  # Exponential backoff\n    # AuthenticationError, BadRequestError etc. → raise immediately\n```\n\n### 4. Prompt Caching\n\nCache long system prompts to avoid resending them on every request.\n\n```python\nmessages = [\n    {\n        \"role\": \"user\",\n        \"content\": [\n            {\n                \"type\": \"text\",\n                \"text\": system_prompt,\n                \"cache_control\": {\"type\": \"ephemeral\"},  # Cache this\n            },\n            {\n                \"type\": \"text\",\n                \"text\": user_input,  # Variable part\n            },\n        ],\n    }\n]\n```\n\n## Composition\n\nCombine all four techniques in a single pipeline function:\n\n```python\ndef process(text: str, config: Config, tracker: CostTracker) -> tuple[Result, CostTracker]:\n    # 1. Route model\n    model = select_model(len(text), estimated_items, config.force_model)\n\n    # 2. Check budget\n    if tracker.over_budget:\n        raise BudgetExceededError(tracker.total_cost, tracker.budget_limit)\n\n    # 3. Call with retry + caching\n    response = call_with_retry(lambda: client.messages.create(\n        model=model,\n        messages=build_cached_messages(system_prompt, text),\n    ))\n\n    # 4. Track cost (immutable)\n    record = CostRecord(model=model, input_tokens=..., output_tokens=..., cost_usd=...)\n    tracker = tracker.add(record)\n\n    return parse_result(response), tracker\n```\n\n## Pricing Reference (2025-2026)\n\n| Model | Input ($\u002F1M tokens) | Output ($\u002F1M tokens) | Relative Cost |\n|-------|---------------------|----------------------|---------------|\n| Haiku 4.5 | $0.80 | $4.00 | 1x |\n| Sonnet 4.6 | $3.00 | $15.00 | ~4x |\n| Opus 4.5 | $15.00 | $75.00 | ~19x |\n\n## Best Practices\n\n- **Start with the cheapest model** and only route to expensive models when complexity thresholds are met\n- **Set explicit budget limits** before processing batches — fail early rather than overspend\n- **Log model selection decisions** so you can tune thresholds based on real data\n- **Use prompt caching** for system prompts over 1024 tokens — saves both cost and latency\n- **Never retry on authentication or validation errors** — only transient failures (network, rate limit, server error)\n\n## Anti-Patterns to Avoid\n\n- Using the most expensive model for all requests regardless of complexity\n- Retrying on all errors (wastes budget on permanent failures)\n- Mutating cost tracking state (makes debugging and auditing difficult)\n- Hardcoding model names throughout the codebase (use constants or config)\n- Ignoring prompt caching for repetitive system prompts\n\n## When to Use\n\n- Any application calling Claude, OpenAI, or similar LLM APIs\n- Batch processing pipelines where cost adds up quickly\n- Multi-model architectures that need intelligent routing\n- Production systems that need budget guardrails\n","# Cost-Aware LLM Pipeline\n\nPatterns for controlling LLM API costs while maintaining quality. Combines model routing, budget tracking, retry logic, and prompt caching into a composable pipeline.\n\n## When to Activate\n\n- Building applications that call LLM APIs (Claude, GPT, etc.)\n- Processing batches of items with varying complexity\n- Need to stay within a budget for API spend\n- Optimizing cost without sacrificing quality on complex tasks\n\n## Core Concepts\n\n### 1. Model Routing by Task Complexity\n\nAutomatically select cheaper models for simple tasks, reserving expensive models for complex ones.\n\n```python\nMODEL_SONNET = \"claude-sonnet-4-6\"\nMODEL_HAIKU = \"claude-haiku-4-5-20251001\"\n\n_SONNET_TEXT_THRESHOLD = 10_000  # chars\n_SONNET_ITEM_THRESHOLD = 30     # items\n\ndef select_model(\n    text_length: int,\n    item_count: int,\n    force_model: str | None = None,\n) -> str:\n    \"\"\"Select model based on task complexity.\"\"\"\n    if force_model is not None:\n        return force_model\n    if text_length >= _SONNET_TEXT_THRESHOLD or item_count >= _SONNET_ITEM_THRESHOLD:\n        return MODEL_SONNET  # Complex task\n    return MODEL_HAIKU  # Simple task (3-4x cheaper)\n```\n\n### 2. Immutable Cost Tracking\n\nTrack cumulative spend with frozen dataclasses. Each API call returns a new tracker — never mutates state.\n\n```python\nfrom dataclasses import dataclass\n\n@dataclass(frozen=True, slots=True)\nclass CostRecord:\n    model: str\n    input_tokens: int\n    output_tokens: int\n    cost_usd: float\n\n@dataclass(frozen=True, slots=True)\nclass CostTracker:\n    budget_limit: float = 1.00\n    records: tuple[CostRecord, ...] = ()\n\n    def add(self, record: CostRecord) -> \"CostTracker\":\n        \"\"\"Return new tracker with added record (never mutates self).\"\"\"\n        return CostTracker(\n            budget_limit=self.budget_limit,\n            records=(*self.records, record),\n        )\n\n    @property\n    def total_cost(self) -> float:\n        return sum(r.cost_usd for r in self.records)\n\n    @property\n    def over_budget(self) -> bool:\n        return self.total_cost > self.budget_limit\n```\n\n### 3. Narrow Retry Logic\n\nRetry only on transient errors. Fail fast on authentication or bad request errors.\n\n```python\nfrom anthropic import (\n    APIConnectionError,\n    InternalServerError,\n    RateLimitError,\n)\n\n_RETRYABLE_ERRORS = (APIConnectionError, RateLimitError, InternalServerError)\n_MAX_RETRIES = 3\n\ndef call_with_retry(func, *, max_retries: int = _MAX_RETRIES):\n    \"\"\"Retry only on transient errors, fail fast on others.\"\"\"\n    for attempt in range(max_retries):\n        try:\n            return func()\n        except _RETRYABLE_ERRORS:\n            if attempt == max_retries - 1:\n                raise\n            time.sleep(2 ** attempt)  # Exponential backoff\n    # AuthenticationError, BadRequestError etc. → raise immediately\n```\n\n### 4. Prompt Caching\n\nCache long system prompts to avoid resending them on every request.\n\n```python\nmessages = [\n    {\n        \"role\": \"user\",\n        \"content\": [\n            {\n                \"type\": \"text\",\n                \"text\": system_prompt,\n                \"cache_control\": {\"type\": \"ephemeral\"},  # Cache this\n            },\n            {\n                \"type\": \"text\",\n                \"text\": user_input,  # Variable part\n            },\n        ],\n    }\n]\n```\n\n## Composition\n\nCombine all four techniques in a single pipeline function:\n\n```python\ndef process(text: str, config: Config, tracker: CostTracker) -> tuple[Result, CostTracker]:\n    # 1. Route model\n    model = select_model(len(text), estimated_items, config.force_model)\n\n    # 2. Check budget\n    if tracker.over_budget:\n        raise BudgetExceededError(tracker.total_cost, tracker.budget_limit)\n\n    # 3. Call with retry + caching\n    response = call_with_retry(lambda: client.messages.create(\n        model=model,\n        messages=build_cached_messages(system_prompt, text),\n    ))\n\n    # 4. Track cost (immutable)\n    record = CostRecord(model=model, input_tokens=..., output_tokens=..., cost_usd=...)\n    tracker = tracker.add(record)\n\n    return parse_result(response), tracker\n```\n\n## Pricing Reference (2025-2026)\n\n| Model | Input ($\u002F1M tokens) | Output ($\u002F1M tokens) | Relative Cost |\n|-------|---------------------|----------------------|---------------|\n| Haiku 4.5 | $0.80 | $4.00 | 1x |\n| Sonnet 4.6 | $3.00 | $15.00 | ~4x |\n| Opus 4.5 | $15.00 | $75.00 | ~19x |\n\n## Best Practices\n\n- **Start with the cheapest model** and only route to expensive models when complexity thresholds are met\n- **Set explicit budget limits** before processing batches — fail early rather than overspend\n- **Log model selection decisions** so you can tune thresholds based on real data\n- **Use prompt caching** for system prompts over 1024 tokens — saves both cost and latency\n- **Never retry on authentication or validation errors** — only transient failures (network, rate limit, server error)\n\n## Anti-Patterns to Avoid\n\n- Using the most expensive model for all requests regardless of complexity\n- Retrying on all errors (wastes budget on permanent failures)\n- Mutating cost tracking state (makes debugging and auditing difficult)\n- Hardcoding model names throughout the codebase (use constants or config)\n- Ignoring prompt caching for repetitive system prompts\n\n## When to Use\n\n- Any application calling Claude, OpenAI, or similar LLM APIs\n- Batch processing pipelines where cost adds up quickly\n- Multi-model architectures that need intelligent routing\n- Production systems that need budget guardrails\n",{"name":5,"origin":48,"description":7},"ECC","master","8ec37df6d71fb664ea19d32c079e7d3ff931a8e0","skills\u002Fecc\u002Fcost-aware-llm-pipeline",3,"https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit","https:\u002F\u002Fweblyfe.ai","https:\u002F\u002Fgithub.com\u002FS3YED","2026-08-27 01:54:02",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add S3YED\u002Fappie-kit --skill cost-aware-llm-pipeline",{"tool":63,"command":64},"bunx","bunx skills add S3YED\u002Fappie-kit --skill cost-aware-llm-pipeline",{"tool":66,"command":67},"pnpm","pnpm dlx skills add S3YED\u002Fappie-kit --skill cost-aware-llm-pipeline","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fecc\u002Fcost-aware-llm-pipeline\u002FSKILL.md",true,[],[72,104,139,165,195,219],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":85,"publishedAt":43},"01M6JFD4M6N4DSGC7TXEJ5R3J2","ai-music","AI music generation and audio tools: HeartMuLa lyrics-to-music generation, songwriting craft (+ Suno prompts), and audio spectrogram\u002Ffeature visualization.","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fwethlete\u002Fai-music\u002FSKILL.md",[78,81,84],{"categoryNo":79,"categoryName":80,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000013","分析与可视化",{"categoryNo":82,"categoryName":83,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103],"music","ai music","humanizer","visualization","Tools","Audio","hermes","hermes agent","songwriting","suno","lyrics","prompts","heartmula","craft","generation","feature","ai music generation","music generation",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":118,"publishedAt":43},"01M6JFCZE4XNFQFMNZDHKHT5V5","agentic-webdesign-design-system-ingestion","Use v0 Design Systems 2.0, Bolt Design System Agents, or Lovable to generate production UI from a client's real design system components — not generic Tailwind. Use when building a client website where brand consistency matters.","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fweb-design\u002Fagentic-webdesign-design-system-ingestion\u002FSKILL.md",[110,113],{"categoryNo":111,"categoryName":112,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":114,"categoryName":115,"taxonomyType":116,"taxonomyTypeDesc":117},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138],"design","agentic","design system","system","ingestion","webdesign","agentic webdesign design system ingestion","components","generate","design systems","website","ui","Tailwind","bolt","lovable","v0","design-system","wed","design system components","appie",{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":147,"publishedAt":43},"01M6JFCVPX3913CFAS6QQJ33ZB","agentic-webdesign-agent-elements","Use 21st.dev Agent Elements — 25 shadcn\u002Fui components purpose-built for AI agent UIs. Tool cards (Bash, Edit, Search, Plan, Subagent, MCP), chat, input, streaming. Open source, shadcn-registry install, no runtime lock-in.","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fweb-design\u002Fagentic-webdesign-agent-elements\u002FSKILL.md",[145,146],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":114,"categoryName":115,"taxonomyType":116,"taxonomyTypeDesc":117},[120,148,124,149,150,151,126,152,153,154,155,156,157,158,159,160,161,162,163,164],"elements","agentic webdesign agent elements","MCP","ai agent","registry","search","streaming","open source","open","Bash","plan","subagent","edit","install","input","lock","source",{"skillNo":166,"name":167,"nameZh":6,"description":168,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":169,"categories":170,"keywords":175,"publishedAt":43},"01M6JFCQC7H65M9QJ6TRE71BS1","vercel-deployment","Deploy static sites and frontends to Vercel — handles both CLI and REST API paths, including the CLI token sanitization workaround.","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Fvercel-deployment\u002FSKILL.md",[171,174],{"categoryNo":172,"categoryName":173,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000011","运维与部署",{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[176,177,178,179,180,181,182,183,184,185,38,186,187,188,189,190,191,192,193,194],"deployment","vercel deployment","vercel","macos","cli","linux","windows","deploy","vercel deploy","REST","site","deploy to vercel","token","rest api","host","sites","paths","sanitization","static site",{"skillNo":196,"name":197,"nameZh":6,"description":198,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":199,"categories":200,"keywords":202,"publishedAt":43},"01M6JFCM4SMTXBM5A01BK2YQG1","software-quality-workflows","Umbrella workflow for TDD, debugging, code simplification, spikes, pre-commit review, and debugger-assisted investigation.","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Fsoftware-quality-workflows\u002FSKILL.md",[201],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,27],"workflows","quality","software","software quality workflows","review","tdd","workflow","commit","debugging","code","debugger","pre-commit","investigation","code simplification","user","simplification",{"skillNo":220,"name":221,"nameZh":6,"description":222,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":223,"categories":224,"keywords":232,"publishedAt":43},"01M6JFCGD0M6CEATY01W7YN6DB","react-component-refactoring","Safely refactor large React components into smaller, maintainable pieces using proven patterns","https:\u002F\u002Fgithub.com\u002FS3YED\u002Fappie-kit\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Freact-component-refactoring\u002FSKILL.md",[225,228,231],{"categoryNo":226,"categoryName":227,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":229,"categoryName":230,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[233,234,235,236,237,238,239,240,179,241,242,126,243,182,244,212,245,246,32,247],"react","component","refactoring","component refactoring","react component","react component refactoring","frontend","writing","safe","typescript","react components","modular","refactor","simplify","test-driven-development"]