[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FAFHR2YSR22YWM37ZCZRNF":3,"skill-same:01M7FAFHR2YSR22YWM37ZCZRNF":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":6,"ownerDescription":35,"updatedAt":36,"install":37,"resources":50},"01M7FAFHR2YSR22YWM37ZCZRNF","prompt-caching-strategy",null,"Optimize LLM prompt caching hit rate to reduce API costs and improve latency. Analyzes prompt structure, identifies cacheable vs non-cacheable content, and recommends restructuring to maximize cache reuse. TRIGGER when: user says \"prompt cache\", \"cache hit rate\", \"reduce LLM cost\", \"optimize token usage\", \"cheaper LLM calls\", or asks about prompt caching strategies for OpenAI\u002FAnthropic\u002Fother providers. DO NOT TRIGGER when: user just wants to optimize a single prompt's quality (use prompt-optimizer instead), or asks about general caching infrastructure.","siddu1004\u002FSFP","siddu1004","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F260113491?v=4","01M7F8W1SAJAR49097VRTE1TW5","Siddhardh Vanguri","01M7F8W1SYGY7JE15AHAECCT9R",5,"MIT","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fai-ml\u002Fskills\u002Fprompt-caching-strategy\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: prompt-caching-strategy\ndescription: >-\n  Optimize LLM prompt caching hit rate to reduce API costs and improve latency.\n  Analyzes prompt structure, identifies cacheable vs non-cacheable content,\n  and recommends restructuring to maximize cache reuse.\n  TRIGGER when: user says \"prompt cache\", \"cache hit rate\", \"reduce LLM cost\",\n  \"optimize token usage\", \"cheaper LLM calls\", or asks about prompt caching\n  strategies for OpenAI\u002FAnthropic\u002Fother providers.\n  DO NOT TRIGGER when: user just wants to optimize a single prompt's quality\n  (use prompt-optimizer instead), or asks about general caching infrastructure.\nmetadata:\n  origin: community\n  author: cyberspace-cs\n  version: \"1.0.0\"\n---\n\n# Prompt Caching Strategy\n\nArrange reusable prompt content to improve cache reuse, then measure input cost\nand latency. Savings depend on model, platform, cache eligibility, writes,\nretention, reuse and output volume. Runtime cache hits and savings for this skill\nare **unmeasured**; the examples below validate arithmetic without provider calls.\n\n## When to Activate\n\nUse for repeated API requests that share authorized instructions, tools or\nreference material. First identify the exact provider endpoint, model, service\ntier and caching mode. Consult its current documentation before selecting rates,\nminimum prefix length or retention. Caching does not reduce context-window use.\n\n## How It Works\n\n### Preserve a stable prefix\n\nKeep reusable instructions, tool schemas and examples stable; place changing\nquestions, timestamps and retrieved results after that content where the API\nallows. Respect the provider's serialization order: Claude uses tools, system,\nthen messages. Matching prefixes need identical content, including tool schemas\nand relevant request settings. A changed prefix can prevent reuse from that\npoint. A dynamic suffix may itself be cached later if reused unchanged.\n\n```text\nReusable prefix: approved tools, instructions, stable examples\u002Freference\nChanging suffix: current question, fresh retrieval results, request metadata\n```\n\nChoose only relevant context. Padding prompts or broadening access just to meet\nan eligibility threshold needs its own cost and quality justification.\n\n### Preserve user and workspace boundaries\n\nAuthorize content access before cache lookup or reuse. Scope application-managed\ncache handles and accounting keys to the approved user, workspace, provider\naccount, model and content revision. Share only material explicitly approved for\nthat scope. A cache key is a routing\u002Faccounting hint, not an authorization gate\nor a sandbox. Provider cache isolation does not replace application access checks.\nInvalidate application references when permissions or content change; follow\nprovider retention\u002Fdeletion controls. Avoid logging private prompt contents.\nContext selection, capabilities, sandbox enforcement and evidence remain\nindependent of cache optimization.\n\n## Provider Requirements and Usage Evidence\n\nThe linked documentation is authoritative; endpoint and SDK versions can expose\ndifferent usage field names. Record that version alongside counters.\n\n| Provider | Requirements to verify | Counters to record |\n| --- | --- | --- |\n| OpenAI | Model-specific automatic\u002Fexplicit modes, minimum length, retention and write\u002Fread prices; earlier and newer models differ | Responses `usage.input_tokens_details.cached_tokens` and, when present, `cache_write_tokens`; input and output totals. Chat Completions uses `prompt_tokens_details.cached_tokens` |\n| Claude | Supported platform, model minimum, `cache_control` support and TTL; cache writes may cost more than ordinary input | `input_tokens`, `cache_creation_input_tokens`, `cache_read_input_tokens`, `output_tokens`; split writes by TTL when rates differ |\n| Gemini | Model minimum and API: Interactions supports implicit caching; explicit cache objects require GenerateContent | Interactions `usage.total_cached_tokens`; GenerateContent `usage_metadata` cache\u002Finput\u002Foutput counters plus explicit-cache token count and retained duration |\n\nUse [OpenAI's prompt caching guide](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Fdocs\u002Fguides\u002Fprompt-caching)\nand [pricing](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Fdocs\u002Fpricing\u002F) for the selected\nmodel. For Chat Completions, check the [usage schema](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Freference\u002Fresources\u002Fchat\u002Fsubresources\u002Fcompletions\u002Fmethods\u002Fcreate).\nDerive ordinary input from the total minus cached reads and any separately\nreported writes; do not count the same tokens twice.\n\nUse [Claude's prompt caching guide](https:\u002F\u002Fplatform.claude.com\u002Fdocs\u002Fen\u002Fbuild-with-claude\u002Fprompt-caching)\nfor current model\u002Fplatform limits and TTL rates. Its `input_tokens` counts\nordinary input separately from cache reads and writes. A short prefix can produce\nzero cache counters even when marked for caching. Do not assume every Claude\nmodel or hosted platform uses the same minimum or read price.\n\nUse [Gemini's caching guide](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fcaching),\n[GenerateContent caching guide](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fgenerate-content\u002Fcaching)\nand [pricing](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fpricing). Explicit-cache\nstorage is charged by token count and retained duration, separately from reads,\nordinary input and output. Implicit hits are not guaranteed. Verify billing\ncategories for the chosen endpoint rather than copying another provider's rates.\n\n## Cost Comparison Calculator\n\nNormalize actual usage into nonoverlapping categories and use prices in the same\ncurrency per million tokens. Split write categories when TTL, modality, tier or\nmodel prices differ. Storage uses million-token-hours; other applicable charges\nsuch as tools and grounding need their own lines. Include all warmup, miss,\nexpiration, refresh and retry requests in the measured period.\n\n```text\nbaseline = (ordinary + written + read) * ordinary_input_rate \u002F 1e6\n           + output * output_rate \u002F 1e6\nactual   = ordinary * ordinary_input_rate \u002F 1e6\n           + written * cache_write_rate \u002F 1e6\n           + read * cache_read_rate \u002F 1e6\n           + output * output_rate \u002F 1e6\n           + storage_token_hours * storage_rate \u002F 1e6\nsavings  = baseline - actual\n```\n\nThis counterfactual holds prompt and output volumes fixed. For explicit-cache\ncreation, include any separately billed creation input in the write category;\nuse the endpoint's documented rate. Set nonexistent categories to zero. Compare\nactual before\u002Fafter invoices separately if warmup requests or output differ.\nNever apply a cache discount to the total bill or to output tokens.\n\nThe executable calculator takes normalized counters, not raw provider responses:\n\n```javascript\nfunction compareCacheCost(usage, rates) {\n  const counts = ['uncached', 'written', 'read', 'output', 'storageTokenHours'];\n  const prices = ['input', 'write', 'read', 'output', 'storage'];\n  for (const [values, keys] of [[usage, counts], [rates, prices]]) {\n    for (const key of keys) {\n      if (!Number.isFinite(values[key]) || values[key] \u003C 0) {\n        throw new Error(`${key} must be a finite nonnegative number`);\n      }\n    }\n  }\n  const outputCost = usage.output * rates.output \u002F 1e6;\n  const storageCost = usage.storageTokenHours * rates.storage \u002F 1e6;\n  const baseline = (usage.uncached + usage.written + usage.read)\n    * rates.input \u002F 1e6 + outputCost;\n  const actual = (usage.uncached * rates.input + usage.written * rates.write\n    + usage.read * rates.read) \u002F 1e6 + outputCost + storageCost;\n  return { baseline, actual, savings: baseline - actual, outputCost, storageCost };\n}\n```\n\n## Examples\n\nThese are hypothetical prices and counts, not provider quotes or measured hits.\nTen calls each contain a 10,000-token stable prefix, 2,000-token changing suffix\nand 2,000 output tokens. Assume one prefix write and nine full prefix reads.\n\n| Category | Tokens | Example rate per million | Cost |\n| --- | ---: | ---: | ---: |\n| Ordinary input | 20,000 | $3.00 | $0.0600 |\n| Cache write | 10,000 | $3.75 | $0.0375 |\n| Cache read | 90,000 | $0.30 | $0.0270 |\n| Output | 20,000 | $15.00 | $0.3000 |\n| Storage | 0 token-hours | $0.00 | $0.0000 |\n\nBaseline: 120,000 input tokens at $3 plus the same output = **$0.6600**.\nWith caching: **$0.4245**, savings **$0.2355**. The write premium is included;\nthe $0.3000 output bill remains unchanged. An output-heavy workload has the same\nabsolute input savings under these assumptions, a smaller fraction of total cost.\n\nIf an explicit cache retains 10,000 tokens for two hours at a hypothetical\n$1 per million-token-hour, add **$0.0200** storage. Actual cost becomes\n**$0.4445**. Use the selected Gemini endpoint's prices for a real estimate.\n\nFor a one-off call, that prefix write costs $0.0375 rather than $0.0300 ordinary\ninput: **$0.0075 more**, with no read savings. For a 500-token prefix below a\nselected model's documented minimum, treat the request as uncached; claiming\nfuture hits without counters would overstate savings. Repeated expiry or changing\nprefixes can also turn expected savings into higher cost.\n\n## Verification Workflow\n\n1. Capture eligible prefix size and model\u002Fplatform requirements from current docs.\n2. Run the offline example and edge cases with\n   `node tests\u002Fskills\u002Fprompt-caching-strategy.test.js` in the ECC repository.\n3. If provider traffic is authorized, compare matched requests and aggregate\n   actual cached-token counts, all input categories, output, storage, latency and\n   realized cost by approved user\u002Fworkspace. Record misses and rewrites too.\n4. Report cached reads divided by total input tokens as a token hit rate; report\n   request hit rate separately. Keep denominators and sampling period explicit.\n5. If no traffic was measured, label hit rate, latency improvement and realized\n   savings unmeasured. Offline arithmetic establishes no runtime cache behavior.\n\n## Relationship to Other Skills\n\n- **prompt-optimizer** improves prompt quality before optimizing reuse.\n- **context-budget** identifies unnecessary context; caching still consumes it.\n- **cost-aware-llm-pipeline** compares caching with routing, retries and other costs.\n","# Prompt Caching Strategy\n\nArrange reusable prompt content to improve cache reuse, then measure input cost\nand latency. Savings depend on model, platform, cache eligibility, writes,\nretention, reuse and output volume. Runtime cache hits and savings for this skill\nare **unmeasured**; the examples below validate arithmetic without provider calls.\n\n## When to Activate\n\nUse for repeated API requests that share authorized instructions, tools or\nreference material. First identify the exact provider endpoint, model, service\ntier and caching mode. Consult its current documentation before selecting rates,\nminimum prefix length or retention. Caching does not reduce context-window use.\n\n## How It Works\n\n### Preserve a stable prefix\n\nKeep reusable instructions, tool schemas and examples stable; place changing\nquestions, timestamps and retrieved results after that content where the API\nallows. Respect the provider's serialization order: Claude uses tools, system,\nthen messages. Matching prefixes need identical content, including tool schemas\nand relevant request settings. A changed prefix can prevent reuse from that\npoint. A dynamic suffix may itself be cached later if reused unchanged.\n\n```text\nReusable prefix: approved tools, instructions, stable examples\u002Freference\nChanging suffix: current question, fresh retrieval results, request metadata\n```\n\nChoose only relevant context. Padding prompts or broadening access just to meet\nan eligibility threshold needs its own cost and quality justification.\n\n### Preserve user and workspace boundaries\n\nAuthorize content access before cache lookup or reuse. Scope application-managed\ncache handles and accounting keys to the approved user, workspace, provider\naccount, model and content revision. Share only material explicitly approved for\nthat scope. A cache key is a routing\u002Faccounting hint, not an authorization gate\nor a sandbox. Provider cache isolation does not replace application access checks.\nInvalidate application references when permissions or content change; follow\nprovider retention\u002Fdeletion controls. Avoid logging private prompt contents.\nContext selection, capabilities, sandbox enforcement and evidence remain\nindependent of cache optimization.\n\n## Provider Requirements and Usage Evidence\n\nThe linked documentation is authoritative; endpoint and SDK versions can expose\ndifferent usage field names. Record that version alongside counters.\n\n| Provider | Requirements to verify | Counters to record |\n| --- | --- | --- |\n| OpenAI | Model-specific automatic\u002Fexplicit modes, minimum length, retention and write\u002Fread prices; earlier and newer models differ | Responses `usage.input_tokens_details.cached_tokens` and, when present, `cache_write_tokens`; input and output totals. Chat Completions uses `prompt_tokens_details.cached_tokens` |\n| Claude | Supported platform, model minimum, `cache_control` support and TTL; cache writes may cost more than ordinary input | `input_tokens`, `cache_creation_input_tokens`, `cache_read_input_tokens`, `output_tokens`; split writes by TTL when rates differ |\n| Gemini | Model minimum and API: Interactions supports implicit caching; explicit cache objects require GenerateContent | Interactions `usage.total_cached_tokens`; GenerateContent `usage_metadata` cache\u002Finput\u002Foutput counters plus explicit-cache token count and retained duration |\n\nUse [OpenAI's prompt caching guide](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Fdocs\u002Fguides\u002Fprompt-caching)\nand [pricing](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Fdocs\u002Fpricing\u002F) for the selected\nmodel. For Chat Completions, check the [usage schema](https:\u002F\u002Fdevelopers.openai.com\u002Fapi\u002Freference\u002Fresources\u002Fchat\u002Fsubresources\u002Fcompletions\u002Fmethods\u002Fcreate).\nDerive ordinary input from the total minus cached reads and any separately\nreported writes; do not count the same tokens twice.\n\nUse [Claude's prompt caching guide](https:\u002F\u002Fplatform.claude.com\u002Fdocs\u002Fen\u002Fbuild-with-claude\u002Fprompt-caching)\nfor current model\u002Fplatform limits and TTL rates. Its `input_tokens` counts\nordinary input separately from cache reads and writes. A short prefix can produce\nzero cache counters even when marked for caching. Do not assume every Claude\nmodel or hosted platform uses the same minimum or read price.\n\nUse [Gemini's caching guide](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fcaching),\n[GenerateContent caching guide](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fgenerate-content\u002Fcaching)\nand [pricing](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fpricing). Explicit-cache\nstorage is charged by token count and retained duration, separately from reads,\nordinary input and output. Implicit hits are not guaranteed. Verify billing\ncategories for the chosen endpoint rather than copying another provider's rates.\n\n## Cost Comparison Calculator\n\nNormalize actual usage into nonoverlapping categories and use prices in the same\ncurrency per million tokens. Split write categories when TTL, modality, tier or\nmodel prices differ. Storage uses million-token-hours; other applicable charges\nsuch as tools and grounding need their own lines. Include all warmup, miss,\nexpiration, refresh and retry requests in the measured period.\n\n```text\nbaseline = (ordinary + written + read) * ordinary_input_rate \u002F 1e6\n           + output * output_rate \u002F 1e6\nactual   = ordinary * ordinary_input_rate \u002F 1e6\n           + written * cache_write_rate \u002F 1e6\n           + read * cache_read_rate \u002F 1e6\n           + output * output_rate \u002F 1e6\n           + storage_token_hours * storage_rate \u002F 1e6\nsavings  = baseline - actual\n```\n\nThis counterfactual holds prompt and output volumes fixed. For explicit-cache\ncreation, include any separately billed creation input in the write category;\nuse the endpoint's documented rate. Set nonexistent categories to zero. Compare\nactual before\u002Fafter invoices separately if warmup requests or output differ.\nNever apply a cache discount to the total bill or to output tokens.\n\nThe executable calculator takes normalized counters, not raw provider responses:\n\n```javascript\nfunction compareCacheCost(usage, rates) {\n  const counts = ['uncached', 'written', 'read', 'output', 'storageTokenHours'];\n  const prices = ['input', 'write', 'read', 'output', 'storage'];\n  for (const [values, keys] of [[usage, counts], [rates, prices]]) {\n    for (const key of keys) {\n      if (!Number.isFinite(values[key]) || values[key] \u003C 0) {\n        throw new Error(`${key} must be a finite nonnegative number`);\n      }\n    }\n  }\n  const outputCost = usage.output * rates.output \u002F 1e6;\n  const storageCost = usage.storageTokenHours * rates.storage \u002F 1e6;\n  const baseline = (usage.uncached + usage.written + usage.read)\n    * rates.input \u002F 1e6 + outputCost;\n  const actual = (usage.uncached * rates.input + usage.written * rates.write\n    + usage.read * rates.read) \u002F 1e6 + outputCost + storageCost;\n  return { baseline, actual, savings: baseline - actual, outputCost, storageCost };\n}\n```\n\n## Examples\n\nThese are hypothetical prices and counts, not provider quotes or measured hits.\nTen calls each contain a 10,000-token stable prefix, 2,000-token changing suffix\nand 2,000 output tokens. Assume one prefix write and nine full prefix reads.\n\n| Category | Tokens | Example rate per million | Cost |\n| --- | ---: | ---: | ---: |\n| Ordinary input | 20,000 | $3.00 | $0.0600 |\n| Cache write | 10,000 | $3.75 | $0.0375 |\n| Cache read | 90,000 | $0.30 | $0.0270 |\n| Output | 20,000 | $15.00 | $0.3000 |\n| Storage | 0 token-hours | $0.00 | $0.0000 |\n\nBaseline: 120,000 input tokens at $3 plus the same output = **$0.6600**.\nWith caching: **$0.4245**, savings **$0.2355**. The write premium is included;\nthe $0.3000 output bill remains unchanged. An output-heavy workload has the same\nabsolute input savings under these assumptions, a smaller fraction of total cost.\n\nIf an explicit cache retains 10,000 tokens for two hours at a hypothetical\n$1 per million-token-hour, add **$0.0200** storage. Actual cost becomes\n**$0.4445**. Use the selected Gemini endpoint's prices for a real estimate.\n\nFor a one-off call, that prefix write costs $0.0375 rather than $0.0300 ordinary\ninput: **$0.0075 more**, with no read savings. For a 500-token prefix below a\nselected model's documented minimum, treat the request as uncached; claiming\nfuture hits without counters would overstate savings. Repeated expiry or changing\nprefixes can also turn expected savings into higher cost.\n\n## Verification Workflow\n\n1. Capture eligible prefix size and model\u002Fplatform requirements from current docs.\n2. Run the offline example and edge cases with\n   `node tests\u002Fskills\u002Fprompt-caching-strategy.test.js` in the ECC repository.\n3. If provider traffic is authorized, compare matched requests and aggregate\n   actual cached-token counts, all input categories, output, storage, latency and\n   realized cost by approved user\u002Fworkspace. Record misses and rewrites too.\n4. Report cached reads divided by total input tokens as a token hit rate; report\n   request hit rate separately. Keep denominators and sampling period explicit.\n5. If no traffic was measured, label hit rate, latency improvement and realized\n   savings unmeasured. Offline arithmetic establishes no runtime cache behavior.\n\n## Relationship to Other Skills\n\n- **prompt-optimizer** improves prompt quality before optimizing reuse.\n- **context-budget** identifies unnecessary context; caching still consumes it.\n- **cost-aware-llm-pipeline** compares caching with routing, retries and other costs.\n",{"name":5,"metadata":24,"description":7},{"author":25,"origin":26,"version":27},"cyberspace-cs","community","1.0.0","main","15fa5201935d2b1d172b4e61c2360c908d876d20","plugins\u002Fai-ml\u002Fskills\u002Fprompt-caching-strategy",0,"https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP","https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002Fsfp-hive","https:\u002F\u002Fgithub.com\u002Fsiddu1004","I am a student at Amrita Vishwa Vidyapeetham pursuing my BTECH degree in Electronics and Electrical Engineering (2024-2028).","2026-11-09 16:04:40",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add siddu1004\u002FSFP --skill prompt-caching-strategy",{"tool":43,"command":44},"bunx","bunx skills add siddu1004\u002FSFP --skill prompt-caching-strategy",{"tool":46,"command":47},"pnpm","pnpm dlx skills add siddu1004\u002FSFP --skill prompt-caching-strategy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fai-ml\u002Fskills\u002Fprompt-caching-strategy\u002FSKILL.md",true,[],[52,59,66,73,80,87],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M7FEEDADZBNBVY19H3XZKCBQ","scientific-schematics","Generates scientific diagram drafts using Nano Banana 2 AI with smart iterative refinement. Uses Gemini 3.7 Flash for quality review. Refines when the review requests improvement, with at most two generations. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways, and complex scientific visualizations.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fscientific-schematics\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7FEE0JS04FE58MST48D0MMC","polyhaven-texture-apply","Apply realistic PBR textures from PolyHaven to any Blender object. Supports metals, wood, concrete, fabric, and more. Trigger when asked to texture an object, apply a material from PolyHaven, make something look like metal\u002Fwood\u002Fmarble, or change object surface.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fpolyhaven-texture-apply\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M7FEDJRWV194TY411WVRS84W","polyhaven-studio-setup","Set up a professional product photography studio in Blender using PolyHaven HDRIs and PBR ground materials. Trigger when asked to create a studio setup, add HDRI lighting, set up product photography, or create a showroom scene.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fpolyhaven-studio-setup\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7FED5H2FGS50XGETMGSM0RV","polyhaven-scene-builder","Build complete product scenes using PolyHaven assets — HDRI environment, textured ground\u002Fpedestal, and optional props from the 3D model library. Trigger when asked to build a product scene, create a showcase setup, add a pedestal, or compose a product shot.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fpolyhaven-scene-builder\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7FECSNZ2CJA7Z8TPE5R844R","polyhaven-material-swap","Cycle through multiple PolyHaven materials on a product and render each variant. Great for showing color\u002Ffinish options. Trigger when asked to show material variants, swap textures, create color options, show finish comparisons, or render product in different materials.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fpolyhaven-material-swap\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7FECBPQ6YWH48PYAMS0C5CE","polyhaven-hdri-showcase","Render the same product across multiple HDRI environments and output a comparison grid. Trigger when asked to compare lighting setups, test different environments, show product in multiple settings, or create an HDRI comparison.","https:\u002F\u002Fgithub.com\u002Fsiddu1004\u002FSFP\u002Fblob\u002Fmain\u002Fplugins\u002Fscience-hardware\u002Fskills\u002Fpolyhaven-hdri-showcase\u002FSKILL.md",[],[]]