[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MKB4A9R6T1HKA0D1KFKADM":3,"skill-same:01M6MKB4A9R6T1HKA0D1KFKADM":76},{"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":28,"publishedAt":41,"linkOnly":42,"markdownContent":43,"markdownBody":44,"frontMatter":45,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":52,"ownerHtmlUrl":53,"ownerBlogUrl":54,"ownerDescription":55,"updatedAt":56,"install":57,"resources":69},"01M6MKB4A9R6T1HKA0D1KFKADM","skill-meta-prompt",null,"Craft better prompts using proven optimization techniques — use when your prompt needs refinement","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1559,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fnyldn\u002Fclaude-octopus\u002Fskills\u002Fskill-meta-prompt\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000031","产品经理","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40],"Meta","prompt","skill meta prompt","meta prompt","Optimization","use","prompts","craft","refinement","true","techniques","when","2026-11-10 10:12:30",false,"---\nname: skill-meta-prompt\ndescription: \"Craft better prompts using proven optimization techniques — use when your prompt needs refinement\"\ndisable-model-invocation: true\n---\n\n> **Host: Codex CLI** — This skill was designed for Claude Code and adapted for Codex.\n> Cross-reference commands use installed skill names in Codex rather than `\u002Focto:*` slash commands.\n> Use the active Codex shell and subagent tools. Do not claim a provider, model, or host subagent is available until the current session exposes it.\n> For host tool equivalents, see `skills\u002Fblocks\u002Fcodex-host-adapter.md`.\n\n\n# Meta-Prompt Generator Skill\n\n## Overview\n\nGenerate well-structured, verifiable prompts for any use case. Applies proven meta-prompting techniques to minimize hallucination and maximize effectiveness.\n\n```\n┌─────────────────────────────────────────────────────────────────────────────┐\n│                       META-PROMPT GENERATION                                 │\n├─────────────────────────────────────────────────────────────────────────────┤\n│                                                                             │\n│  Phase 1: Requirement Gathering                                             │\n│       → Understand the primary goal\u002Frole                                    │\n│       → Clarify expected outputs                                            │\n│       → Identify accuracy requirements                                      │\n│       ↓                                                                     │\n│  Phase 2: Task Analysis                                                     │\n│       → Apply Technique 1: Task Decomposition                               │\n│       → Identify if complex enough for subtasks                             │\n│       → Map dependencies between subtasks                                   │\n│       ↓                                                                     │\n│  Phase 3: Expert Assignment                                                 │\n│       → Apply Technique 5: Specialized Experts                              │\n│       → Assign personas to subtasks                                         │\n│       → Apply Technique 2: Fresh Eyes Review                                │\n│       ↓                                                                     │\n│  Phase 4: Verification Design                                               │\n│       → Apply Technique 3: Iterative Verification                           │\n│       → Build in checking steps                                             │\n│       → Apply Technique 4: No Guessing                                      │\n│       ↓                                                                     │\n│  Phase 5: Prompt Assembly                                                   │\n│       → Structure: Role, Context, Instructions, Constraints, Format         │\n│       → Add verification hooks                                              │\n│       → Include uncertainty disclaimers                                     │\n│       ↓                                                                     │\n│  Phase 6: Output & Iteration                                                │\n│       → Present generated prompt                                            │\n│       → Offer refinement                                                    │\n│                                                                             │\n└─────────────────────────────────────────────────────────────────────────────┘\n```\n\n\n## The Five Techniques\n\n### Technique 1: Task Decomposition\n\n**What:** Break complex tasks into smaller, manageable subtasks.\n\n**When to use:**\n- Task has multiple distinct steps\n- Different expertise needed for different parts\n- Risk of getting lost in complexity\n\n**How to apply:**\n1. List all components of the task\n2. Identify dependencies (what must happen first)\n3. Group related components\n4. Order by logical sequence\n\n**Example:**\n```\nTask: \"Create a technical blog post about OAuth 2.0\"\n\nDecomposition:\n1. Research Phase\n   - Gather OAuth 2.0 specifications\n   - Find common implementation examples\n   - Identify security best practices\n   \n2. Structure Phase\n   - Outline main sections\n   - Plan code examples\n   - Design diagrams\u002Fvisuals\n   \n3. Writing Phase\n   - Write introduction\n   - Write technical sections\n   - Write conclusion\u002FCTA\n   \n4. Review Phase\n   - Technical accuracy check\n   - Code example testing\n   - Readability review\n```\n\n\n### Technique 2: Fresh Eyes Review\n\n**What:** Use different \"experts\" for creation vs. validation. Never use the same expert to both create and verify.\n\n**When to use:**\n- Output needs to be accurate\n- Risk of blind spots from creator\n- Quality assurance is critical\n\n**How to apply:**\n1. Assign Creator Expert for initial work\n2. Assign different Reviewer Expert for validation\n3. Reviewer should not have seen creation process\n4. Loop back to Creator if issues found\n\n**Example:**\n```\nCreator: \"Expert Technical Writer\" produces article\nReviewer: \"Expert Security Engineer\" verifies OAuth claims\nReviewer: \"Expert Developer\" tests code examples\n\nNOT: Same expert writes AND reviews their own work\n```\n\n\n### Technique 3: Iterative Verification\n\n**What:** Build explicit verification steps into the task, especially for error-prone outputs.\n\n**When to use:**\n- Mathematical calculations\n- Code generation\n- Factual claims\n- Multi-step reasoning\n\n**How to apply:**\n1. After each significant output, add verification step\n2. For calculations: \"Now verify this by [alternative method]\"\n3. For code: \"Test this code against [test cases]\"\n4. For claims: \"Confirm this by [citing source]\"\n\n**Example:**\n```\nStep 1: Calculate discount price\nStep 2: VERIFY - recalculate from opposite direction\nStep 3: If mismatch, identify error and recalculate\nStep 4: Only proceed when both methods match\n```\n\n\n### Technique 4: No Guessing\n\n**What:** Never assume unverified facts. Disclaim uncertainty explicitly.\n\n**When to use:**\n- ALWAYS (this is a default behavior)\n- Especially for: dates, statistics, quotes, technical specifications\n\n**How to apply:**\n1. If uncertain, say \"I'm not certain about...\"\n2. If no data, say \"I don't have information on...\"\n3. Ask for sources rather than inventing\n4. Distinguish between \"likely\" and \"confirmed\"\n\n**Disclaimer templates:**\n```\n\"Note: This figure is approximate and should be verified.\"\n\"I don't have access to [specific data]. Please provide or verify.\"\n\"This is based on general patterns; your specific case may differ.\"\n```\n\n\n### Technique 5: Specialized Experts\n\n**What:** Spawn domain-specific personas for complex subtasks.\n\n**When to use:**\n- Task requires specialized knowledge\n- Different perspectives would improve quality\n- Cross-functional work needed\n\n**Available expert archetypes:**\n| Expert | Use For |\n|--------|---------|\n| Expert Writer | Content, copy, documentation |\n| Expert Mathematician | Calculations, proofs, statistics |\n| Expert Python | Python code, data analysis |\n| Expert Security | Security review, threat modeling |\n| Expert Architect | System design, trade-offs |\n| Expert Reviewer | Quality assurance, error-finding |\n| Expert Strategist | Planning, prioritization |\n\n**How to apply:**\n```\n\"For this subtask, adopt the persona of Expert [X].\nYour expertise includes [specific areas].\nFocus exclusively on [your assigned task].\nYou have no memory of previous context—all needed information is below.\"\n```\n\n\n## Phase 1: Requirement Gathering\n\n### Initial Prompt\n\n```markdown\n**Meta-Prompt Generator**\n\nI'll help you create an effective, verifiable prompt.\n\n**Questions:**\n\n1. **What is the main goal?**\n   What should this prompt help someone accomplish?\n\n2. **What's the expected output?**\n   (e.g., document, code, analysis, decision)\n\n3. **How important is accuracy?**\n   - Critical (factual, technical, or high-stakes)\n   - Moderate (useful but not mission-critical)\n   - Flexible (creative, exploratory)\n\n4. **Any specific constraints?**\n   (length, format, tone, tools available)\n```\n\n### Minimum Information Needed\n\n- Primary goal (REQUIRED)\n- Output type (REQUIRED)\n- Accuracy requirements (can assume moderate)\n- Constraints (optional, will use sensible defaults)\n\n**If information is missing, ask ONE clarifying question at a time.**\n\n\n## Phase 2-4: Analysis & Design\n\nAfter gathering requirements, analyze internally:\n\n### Task Complexity Assessment\n\n| Complexity | Indicators | Approach |\n|------------|------------|----------|\n| **Simple** | Single step, one output | Direct prompt, no decomposition |\n| **Moderate** | 2-3 steps, clear sequence | Light decomposition, one expert |\n| **Complex** | 4+ steps, dependencies | Full decomposition, multiple experts |\n\n### Expert Assignment Matrix\n\n| Task Type | Creator Expert | Reviewer Expert |\n|-----------|----------------|-----------------|\n| Technical writing | Expert Writer | Expert Engineer |\n| Code generation | Expert Developer | Expert Reviewer |\n| Analysis | Expert Analyst | Expert Strategist |\n| Creative | Expert Creative | Expert Editor |\n\n### Verification Points\n\nFor the task, identify where verification is needed:\n\n| Step | Risk | Verification Method |\n|------|------|---------------------|\n| [step] | [what could go wrong] | [how to verify] |\n\n\n## Phase 5: Prompt Assembly\n\n### Output Format\n\nYou MUST return the generated prompt in this exact format:\n\n```markdown\n# [Prompt Title]\n\n## Role\n[Short, direct role definition]\n[Emphasize verification and uncertainty disclaimers]\n\n## Context\n[User's task and goals]\n[Background information provided]\n[Clarifications gathered]\n\n## Instructions\n\n### Phase 1: [First Phase Name]\n1. [Step 1]\n2. [Step 2]\n3. **Verification:** [How to verify this phase]\n\n### Phase 2: [Second Phase Name]\n1. [Step 1]\n2. [Step 2]\n3. **Verification:** [How to verify this phase]\n\n[Continue phases as needed...]\n\n### Expert Assignments (if applicable)\n- **[Expert Type]:** Handles [specific subtask]\n- **[Reviewer Type]:** Validates [what they check]\n\n## Constraints\n- [Constraint 1]\n- [Constraint 2]\n- [Accuracy requirement: how to handle uncertainty]\n\n## Output Format\n[Specify exactly how the output should be structured]\n[Include all required sections]\n\n## Verification Checklist\nBefore considering complete:\n- [ ] [Verification item 1]\n- [ ] [Verification item 2]\n- [ ] [Accuracy disclaimers added where needed]\n\n## Examples (if provided)\n[Context or examples from user]\n```\n\n### Model-Specific Adjustments\n\nTune the assembled prompt to the model that will execute it:\n\n- **Current frontier roster**: apply `skills\u002Fblocks\u002Ffrontier-model-routing.md` when choosing between Opus 5, GPT-5.6, Sonnet 5, Fable 5.1, Astra, or cheaper seats.\n- **Claude Fable 5.1 or 5**: apply `skills\u002Fblocks\u002Ffable5-prompting.md` to either model ID. In short: never instruct the model to reveal or transcribe its reasoning; replace step-by-step micromanagement with a boundary plus checkable acceptance criteria; drop \"CRITICAL\"\u002F\"MUST\" emphasis unless strict compliance is required; add grounded-progress and act-when-ready language for long runs.\n- **GPT-6 Astra**: reserve it for a bounded pass after GPT-5.6 Sol fails a hard acceptance test. Do not treat Astra plus GPT-5.6 as independent review.\n- **Codex \u002F GPT-5.6**: apply `docs\u002FGPT-5.6-PROMPTING.md`; give it a concrete outcome, scoped repository constraints, checkable acceptance criteria, non-goals, and the required verification.\n- **Older Claude models**: the fuller template below applies as written.\n\n\n## Phase 6: Output & Iteration\n\n### Present the Prompt\n\n```markdown\n**Generated Meta-Prompt**\n\n[The prompt in the format above]\n\n\n**Techniques Applied:**\n- ✓ Task Decomposition: [Yes\u002FNo, how]\n- ✓ Fresh Eyes Review: [Yes\u002FNo, which experts]\n- ✓ Iterative Verification: [Yes\u002FNo, where]\n- ✓ No Guessing: [Where disclaimers needed]\n- ✓ Specialized Experts: [Which ones assigned]\n\n**Would you like me to:**\n1. Refine any section\n2. Add more verification steps\n3. Adjust the complexity level\n4. Execute this prompt now\n```\n\n\n## Example: Complete Meta-Prompt Generation\n\n### User Request\n\"Create a prompt to help me write API documentation\"\n\n### Generated Prompt\n\n```markdown\n# API Documentation Generator\n\n## Role\nYou are an Expert Technical Writer specializing in API documentation.\nYour documentation will be reviewed by an Expert Developer for accuracy.\nWhen uncertain about behavior, note \"Verify with implementation\" rather than guessing.\n\n## Context\nCreating comprehensive API documentation for developers.\nDocumentation should be usable without reading source code.\n\n## Instructions\n\n### Phase 1: API Discovery\n1. List all endpoints with methods (GET, POST, etc.)\n2. Document request\u002Fresponse schemas\n3. Identify authentication requirements\n4. **Verification:** Cross-reference with OpenAPI spec if available\n\n### Phase 2: Documentation Writing\n1. Write endpoint descriptions (what it does, not how)\n2. Create request examples with all parameters\n3. Create response examples for success and error cases\n4. Document rate limits and constraints\n5. **Verification:** Each example should be valid JSON\u002Fcode\n\n### Phase 3: Review Cycle\nExpert Developer reviews for:\n- Technical accuracy of examples\n- Missing edge cases\n- Unclear descriptions\n\n### Expert Assignments\n- **Expert Technical Writer:** Creates documentation prose\n- **Expert Developer:** Validates examples and accuracy\n\n## Constraints\n- Use consistent terminology throughout\n- Examples must be syntactically valid\n- Note any undocumented or unclear behaviors\n- Accuracy: Mark assumptions with \"Assumed behavior - verify\"\n\n## Output Format\n```markdown\n# [Endpoint Name]\n\n**Method:** [HTTP method]\n**Path:** [\u002Fapi\u002Fpath]\n**Auth:** [Required\u002FOptional\u002FNone]\n\n## Description\n[What this endpoint does]\n\n## Request\n[Parameters, body schema, headers]\n\n## Response\n[Success and error responses with examples]\n\n## Notes\n[Rate limits, deprecation, related endpoints]\n```\n\n## Verification Checklist\n- [ ] All endpoints documented\n- [ ] All examples are valid\n- [ ] Authentication clearly specified\n- [ ] Error responses included\n- [ ] Assumptions marked for verification\n```\n\n\n## Error Handling\n\n### Unclear Requirements\n\n```markdown\nI need a bit more clarity to create an effective prompt.\n\n**Specifically:**\n[Question about the unclear part]\n\n[Offer 2-3 options if applicable]\n```\n\n### Over-Complex Request\n\n```markdown\nThis task has [N] distinct components. I recommend:\n\n1. **Split into multiple prompts** - One per major component\n2. **Simplify scope** - Focus on [core element] first\n3. **Proceed as-is** - Full complexity, longer prompt\n\nWhich approach works best for you?\n```\n\n### Can't Apply Techniques\n\nIf techniques don't fit the task:\n\n```markdown\nℹ️ **Note on Techniques**\n\nThis task is straightforward enough that some techniques\ndon't apply:\n\n- Task Decomposition: Not needed (single step)\n- Fresh Eyes: [Explain why\u002Fwhy not]\n- Specialized Experts: Not needed (single domain)\n\nThe generated prompt focuses on clarity and verification instead.\n```\n\n\n## Integration\n\n### With skill-content-pipeline\nGenerate prompts for content creation based on anatomy guides.\n\n### With skill-thought-partner\nTransform brainstorming insights into actionable prompts.\n\n### With skill-prd\nEnhance PRD generation with meta-prompting techniques.\n\n### With flow-develop\nGenerate implementation prompts with built-in verification.\n\n\n## The Bottom Line\n\n```\nMeta-prompt → Decompose → Assign experts → Build verification → Generate\nOtherwise → Vague prompts → Hallucination → Unreliable output\n```\n\n**Structure breeds reliability. Verification breeds accuracy. Experts breed quality.**\n","> **Host: Codex CLI** — This skill was designed for Claude Code and adapted for Codex.\n> Cross-reference commands use installed skill names in Codex rather than `\u002Focto:*` slash commands.\n> Use the active Codex shell and subagent tools. Do not claim a provider, model, or host subagent is available until the current session exposes it.\n> For host tool equivalents, see `skills\u002Fblocks\u002Fcodex-host-adapter.md`.\n\n\n# Meta-Prompt Generator Skill\n\n## Overview\n\nGenerate well-structured, verifiable prompts for any use case. Applies proven meta-prompting techniques to minimize hallucination and maximize effectiveness.\n\n```\n┌─────────────────────────────────────────────────────────────────────────────┐\n│                       META-PROMPT GENERATION                                 │\n├─────────────────────────────────────────────────────────────────────────────┤\n│                                                                             │\n│  Phase 1: Requirement Gathering                                             │\n│       → Understand the primary goal\u002Frole                                    │\n│       → Clarify expected outputs                                            │\n│       → Identify accuracy requirements                                      │\n│       ↓                                                                     │\n│  Phase 2: Task Analysis                                                     │\n│       → Apply Technique 1: Task Decomposition                               │\n│       → Identify if complex enough for subtasks                             │\n│       → Map dependencies between subtasks                                   │\n│       ↓                                                                     │\n│  Phase 3: Expert Assignment                                                 │\n│       → Apply Technique 5: Specialized Experts                              │\n│       → Assign personas to subtasks                                         │\n│       → Apply Technique 2: Fresh Eyes Review                                │\n│       ↓                                                                     │\n│  Phase 4: Verification Design                                               │\n│       → Apply Technique 3: Iterative Verification                           │\n│       → Build in checking steps                                             │\n│       → Apply Technique 4: No Guessing                                      │\n│       ↓                                                                     │\n│  Phase 5: Prompt Assembly                                                   │\n│       → Structure: Role, Context, Instructions, Constraints, Format         │\n│       → Add verification hooks                                              │\n│       → Include uncertainty disclaimers                                     │\n│       ↓                                                                     │\n│  Phase 6: Output & Iteration                                                │\n│       → Present generated prompt                                            │\n│       → Offer refinement                                                    │\n│                                                                             │\n└─────────────────────────────────────────────────────────────────────────────┘\n```\n\n\n## The Five Techniques\n\n### Technique 1: Task Decomposition\n\n**What:** Break complex tasks into smaller, manageable subtasks.\n\n**When to use:**\n- Task has multiple distinct steps\n- Different expertise needed for different parts\n- Risk of getting lost in complexity\n\n**How to apply:**\n1. List all components of the task\n2. Identify dependencies (what must happen first)\n3. Group related components\n4. Order by logical sequence\n\n**Example:**\n```\nTask: \"Create a technical blog post about OAuth 2.0\"\n\nDecomposition:\n1. Research Phase\n   - Gather OAuth 2.0 specifications\n   - Find common implementation examples\n   - Identify security best practices\n   \n2. Structure Phase\n   - Outline main sections\n   - Plan code examples\n   - Design diagrams\u002Fvisuals\n   \n3. Writing Phase\n   - Write introduction\n   - Write technical sections\n   - Write conclusion\u002FCTA\n   \n4. Review Phase\n   - Technical accuracy check\n   - Code example testing\n   - Readability review\n```\n\n\n### Technique 2: Fresh Eyes Review\n\n**What:** Use different \"experts\" for creation vs. validation. Never use the same expert to both create and verify.\n\n**When to use:**\n- Output needs to be accurate\n- Risk of blind spots from creator\n- Quality assurance is critical\n\n**How to apply:**\n1. Assign Creator Expert for initial work\n2. Assign different Reviewer Expert for validation\n3. Reviewer should not have seen creation process\n4. Loop back to Creator if issues found\n\n**Example:**\n```\nCreator: \"Expert Technical Writer\" produces article\nReviewer: \"Expert Security Engineer\" verifies OAuth claims\nReviewer: \"Expert Developer\" tests code examples\n\nNOT: Same expert writes AND reviews their own work\n```\n\n\n### Technique 3: Iterative Verification\n\n**What:** Build explicit verification steps into the task, especially for error-prone outputs.\n\n**When to use:**\n- Mathematical calculations\n- Code generation\n- Factual claims\n- Multi-step reasoning\n\n**How to apply:**\n1. After each significant output, add verification step\n2. For calculations: \"Now verify this by [alternative method]\"\n3. For code: \"Test this code against [test cases]\"\n4. For claims: \"Confirm this by [citing source]\"\n\n**Example:**\n```\nStep 1: Calculate discount price\nStep 2: VERIFY - recalculate from opposite direction\nStep 3: If mismatch, identify error and recalculate\nStep 4: Only proceed when both methods match\n```\n\n\n### Technique 4: No Guessing\n\n**What:** Never assume unverified facts. Disclaim uncertainty explicitly.\n\n**When to use:**\n- ALWAYS (this is a default behavior)\n- Especially for: dates, statistics, quotes, technical specifications\n\n**How to apply:**\n1. If uncertain, say \"I'm not certain about...\"\n2. If no data, say \"I don't have information on...\"\n3. Ask for sources rather than inventing\n4. Distinguish between \"likely\" and \"confirmed\"\n\n**Disclaimer templates:**\n```\n\"Note: This figure is approximate and should be verified.\"\n\"I don't have access to [specific data]. Please provide or verify.\"\n\"This is based on general patterns; your specific case may differ.\"\n```\n\n\n### Technique 5: Specialized Experts\n\n**What:** Spawn domain-specific personas for complex subtasks.\n\n**When to use:**\n- Task requires specialized knowledge\n- Different perspectives would improve quality\n- Cross-functional work needed\n\n**Available expert archetypes:**\n| Expert | Use For |\n|--------|---------|\n| Expert Writer | Content, copy, documentation |\n| Expert Mathematician | Calculations, proofs, statistics |\n| Expert Python | Python code, data analysis |\n| Expert Security | Security review, threat modeling |\n| Expert Architect | System design, trade-offs |\n| Expert Reviewer | Quality assurance, error-finding |\n| Expert Strategist | Planning, prioritization |\n\n**How to apply:**\n```\n\"For this subtask, adopt the persona of Expert [X].\nYour expertise includes [specific areas].\nFocus exclusively on [your assigned task].\nYou have no memory of previous context—all needed information is below.\"\n```\n\n\n## Phase 1: Requirement Gathering\n\n### Initial Prompt\n\n```markdown\n**Meta-Prompt Generator**\n\nI'll help you create an effective, verifiable prompt.\n\n**Questions:**\n\n1. **What is the main goal?**\n   What should this prompt help someone accomplish?\n\n2. **What's the expected output?**\n   (e.g., document, code, analysis, decision)\n\n3. **How important is accuracy?**\n   - Critical (factual, technical, or high-stakes)\n   - Moderate (useful but not mission-critical)\n   - Flexible (creative, exploratory)\n\n4. **Any specific constraints?**\n   (length, format, tone, tools available)\n```\n\n### Minimum Information Needed\n\n- Primary goal (REQUIRED)\n- Output type (REQUIRED)\n- Accuracy requirements (can assume moderate)\n- Constraints (optional, will use sensible defaults)\n\n**If information is missing, ask ONE clarifying question at a time.**\n\n\n## Phase 2-4: Analysis & Design\n\nAfter gathering requirements, analyze internally:\n\n### Task Complexity Assessment\n\n| Complexity | Indicators | Approach |\n|------------|------------|----------|\n| **Simple** | Single step, one output | Direct prompt, no decomposition |\n| **Moderate** | 2-3 steps, clear sequence | Light decomposition, one expert |\n| **Complex** | 4+ steps, dependencies | Full decomposition, multiple experts |\n\n### Expert Assignment Matrix\n\n| Task Type | Creator Expert | Reviewer Expert |\n|-----------|----------------|-----------------|\n| Technical writing | Expert Writer | Expert Engineer |\n| Code generation | Expert Developer | Expert Reviewer |\n| Analysis | Expert Analyst | Expert Strategist |\n| Creative | Expert Creative | Expert Editor |\n\n### Verification Points\n\nFor the task, identify where verification is needed:\n\n| Step | Risk | Verification Method |\n|------|------|---------------------|\n| [step] | [what could go wrong] | [how to verify] |\n\n\n## Phase 5: Prompt Assembly\n\n### Output Format\n\nYou MUST return the generated prompt in this exact format:\n\n```markdown\n# [Prompt Title]\n\n## Role\n[Short, direct role definition]\n[Emphasize verification and uncertainty disclaimers]\n\n## Context\n[User's task and goals]\n[Background information provided]\n[Clarifications gathered]\n\n## Instructions\n\n### Phase 1: [First Phase Name]\n1. [Step 1]\n2. [Step 2]\n3. **Verification:** [How to verify this phase]\n\n### Phase 2: [Second Phase Name]\n1. [Step 1]\n2. [Step 2]\n3. **Verification:** [How to verify this phase]\n\n[Continue phases as needed...]\n\n### Expert Assignments (if applicable)\n- **[Expert Type]:** Handles [specific subtask]\n- **[Reviewer Type]:** Validates [what they check]\n\n## Constraints\n- [Constraint 1]\n- [Constraint 2]\n- [Accuracy requirement: how to handle uncertainty]\n\n## Output Format\n[Specify exactly how the output should be structured]\n[Include all required sections]\n\n## Verification Checklist\nBefore considering complete:\n- [ ] [Verification item 1]\n- [ ] [Verification item 2]\n- [ ] [Accuracy disclaimers added where needed]\n\n## Examples (if provided)\n[Context or examples from user]\n```\n\n### Model-Specific Adjustments\n\nTune the assembled prompt to the model that will execute it:\n\n- **Current frontier roster**: apply `skills\u002Fblocks\u002Ffrontier-model-routing.md` when choosing between Opus 5, GPT-5.6, Sonnet 5, Fable 5.1, Astra, or cheaper seats.\n- **Claude Fable 5.1 or 5**: apply `skills\u002Fblocks\u002Ffable5-prompting.md` to either model ID. In short: never instruct the model to reveal or transcribe its reasoning; replace step-by-step micromanagement with a boundary plus checkable acceptance criteria; drop \"CRITICAL\"\u002F\"MUST\" emphasis unless strict compliance is required; add grounded-progress and act-when-ready language for long runs.\n- **GPT-6 Astra**: reserve it for a bounded pass after GPT-5.6 Sol fails a hard acceptance test. Do not treat Astra plus GPT-5.6 as independent review.\n- **Codex \u002F GPT-5.6**: apply `docs\u002FGPT-5.6-PROMPTING.md`; give it a concrete outcome, scoped repository constraints, checkable acceptance criteria, non-goals, and the required verification.\n- **Older Claude models**: the fuller template below applies as written.\n\n\n## Phase 6: Output & Iteration\n\n### Present the Prompt\n\n```markdown\n**Generated Meta-Prompt**\n\n[The prompt in the format above]\n\n\n**Techniques Applied:**\n- ✓ Task Decomposition: [Yes\u002FNo, how]\n- ✓ Fresh Eyes Review: [Yes\u002FNo, which experts]\n- ✓ Iterative Verification: [Yes\u002FNo, where]\n- ✓ No Guessing: [Where disclaimers needed]\n- ✓ Specialized Experts: [Which ones assigned]\n\n**Would you like me to:**\n1. Refine any section\n2. Add more verification steps\n3. Adjust the complexity level\n4. Execute this prompt now\n```\n\n\n## Example: Complete Meta-Prompt Generation\n\n### User Request\n\"Create a prompt to help me write API documentation\"\n\n### Generated Prompt\n\n```markdown\n# API Documentation Generator\n\n## Role\nYou are an Expert Technical Writer specializing in API documentation.\nYour documentation will be reviewed by an Expert Developer for accuracy.\nWhen uncertain about behavior, note \"Verify with implementation\" rather than guessing.\n\n## Context\nCreating comprehensive API documentation for developers.\nDocumentation should be usable without reading source code.\n\n## Instructions\n\n### Phase 1: API Discovery\n1. List all endpoints with methods (GET, POST, etc.)\n2. Document request\u002Fresponse schemas\n3. Identify authentication requirements\n4. **Verification:** Cross-reference with OpenAPI spec if available\n\n### Phase 2: Documentation Writing\n1. Write endpoint descriptions (what it does, not how)\n2. Create request examples with all parameters\n3. Create response examples for success and error cases\n4. Document rate limits and constraints\n5. **Verification:** Each example should be valid JSON\u002Fcode\n\n### Phase 3: Review Cycle\nExpert Developer reviews for:\n- Technical accuracy of examples\n- Missing edge cases\n- Unclear descriptions\n\n### Expert Assignments\n- **Expert Technical Writer:** Creates documentation prose\n- **Expert Developer:** Validates examples and accuracy\n\n## Constraints\n- Use consistent terminology throughout\n- Examples must be syntactically valid\n- Note any undocumented or unclear behaviors\n- Accuracy: Mark assumptions with \"Assumed behavior - verify\"\n\n## Output Format\n```markdown\n# [Endpoint Name]\n\n**Method:** [HTTP method]\n**Path:** [\u002Fapi\u002Fpath]\n**Auth:** [Required\u002FOptional\u002FNone]\n\n## Description\n[What this endpoint does]\n\n## Request\n[Parameters, body schema, headers]\n\n## Response\n[Success and error responses with examples]\n\n## Notes\n[Rate limits, deprecation, related endpoints]\n```\n\n## Verification Checklist\n- [ ] All endpoints documented\n- [ ] All examples are valid\n- [ ] Authentication clearly specified\n- [ ] Error responses included\n- [ ] Assumptions marked for verification\n```\n\n\n## Error Handling\n\n### Unclear Requirements\n\n```markdown\nI need a bit more clarity to create an effective prompt.\n\n**Specifically:**\n[Question about the unclear part]\n\n[Offer 2-3 options if applicable]\n```\n\n### Over-Complex Request\n\n```markdown\nThis task has [N] distinct components. I recommend:\n\n1. **Split into multiple prompts** - One per major component\n2. **Simplify scope** - Focus on [core element] first\n3. **Proceed as-is** - Full complexity, longer prompt\n\nWhich approach works best for you?\n```\n\n### Can't Apply Techniques\n\nIf techniques don't fit the task:\n\n```markdown\nℹ️ **Note on Techniques**\n\nThis task is straightforward enough that some techniques\ndon't apply:\n\n- Task Decomposition: Not needed (single step)\n- Fresh Eyes: [Explain why\u002Fwhy not]\n- Specialized Experts: Not needed (single domain)\n\nThe generated prompt focuses on clarity and verification instead.\n```\n\n\n## Integration\n\n### With skill-content-pipeline\nGenerate prompts for content creation based on anatomy guides.\n\n### With skill-thought-partner\nTransform brainstorming insights into actionable prompts.\n\n### With skill-prd\nEnhance PRD generation with meta-prompting techniques.\n\n### With flow-develop\nGenerate implementation prompts with built-in verification.\n\n\n## The Bottom Line\n\n```\nMeta-prompt → Decompose → Assign experts → Build verification → Generate\nOtherwise → Vague prompts → Hallucination → Unreliable output\n```\n\n**Structure breeds reliability. Verification breeds accuracy. Experts breed quality.**\n",{"name":5,"description":7,"disable-model-invocation":46},true,"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fnyldn\u002Fclaude-octopus\u002Fskills\u002Fskill-meta-prompt",474,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-11 14:40:14",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":46},[59,62,65],{"tool":60,"command":61},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill skill-meta-prompt",{"tool":63,"command":64},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill skill-meta-prompt",{"tool":66,"command":67},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill skill-meta-prompt","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fnyldn\u002Fclaude-octopus\u002Fskills\u002Fskill-meta-prompt\u002FSKILL.md",[70],{"path":71,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":74,"sourceHtmlUrl":75},"agents\u002Fopenai.yaml","REFERENCE","参考文档",172,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fnyldn\u002Fclaude-octopus\u002Fskills\u002Fskill-meta-prompt\u002Fagents\u002Fopenai.yaml",[77,85,92,99,125,150],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":84},"01M746ZWACVKMZ1BJXJZHYJSSX","migrate-to-e2e","Moves an existing Playwright, Cypress, Selenium, Appium, Detox, Maestro, or TesterArmy suite to e2e one slice at a time, with the old suite running beside the new one until each ported test is proven. Keeps a MIGRATION.md progress file so any later session resumes where the last stopped. Optionally plants a defect to show both suites catch it. Use for \u002Fmigrate-to-e2e, \"migrate our tests to e2e\", \"port these Maestro flows\", \"move our TesterArmy tests\", or when a project has an e2e config and an older suite still running.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fmigrate-to-e2e\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":84},"01M746ZPTAX1Q0FBKAB1CYND7E","e2e","Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":84},"01M746ZJN784HVHMDP9MQGCSGG","create-verification-skill","Generate a project-local verify-\u003Capp> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for \u002Fcreate-verification-skill, \"make a verification skill for this repo\", or when a project has no scripted way to prove UI behavior.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fcreate-verification-skill\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":41},"01M716A6ETNB9B1SQ3GXZC75VR","deja-search","deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-search\u002FSKILL.md",[],[106,107,108,109,110,111,112,113,114,115,116,40,117,118,119,120,121,122,123,124],"search","deja search","never","error","ready","memory","re","decide","fix","design","debugging","you","done","reviewer","io","change","cli","merge","past",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":41},"01M716A3J2W5SENSE9VGMXB375","deja-reflect","End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-reflect\u002FSKILL.md",[],[132,133,134,120,135,136,137,138,139,140,141,142,143,144,111,145,146,147,148,149],"deja reflect","reflect","review","with","record","github","next","https","node","Rules","decisions","brew","remember","project rules","project","session","end","reflection",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":163,"publishedAt":41},"01M715T3RR4BAYSR57HPPTJ0PF","revid","Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftmakersas\u002Frevid-skills\u002Fskills\u002Frevid\u002FSKILL.md",[156,159,160],{"categoryNo":157,"categoryName":158,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":161,"categoryName":162,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[152,164,165,166,167,168,169,146,170,112,171,172,173,174,40,175,176,177,178,179],"URL","Video","short-form","standalone","PDF","viral","tiktok","short","social","MCP","song","image","mcp server","schedule","api","publish"]