meta-prompt
Meta-prompting framework for critiquing responses, analyzing solution trajectories, and evaluating AI-generated content quality
What this skill does
# Meta-Prompt A collection of meta-prompting techniques for evaluating and analyzing AI responses and solution paths. ## Response Quality Evaluator A framework for critiquing and reflecting on the quality of responses, providing a score and indicating whether the response has fully solved the question or task. ### Evaluation Fields **Reflections**: The critique and reflections on the sufficiency, superfluency, and general quality of the response. **Score**: Score from 0-10 on the quality of the candidate response. **Found_solution**: Whether the response has fully solved the question or task. ### Evaluation Criteria When evaluating responses, consider the following: 1. **Accuracy**: Does the response correctly address the question or task? 2. **Completeness**: Does it cover all aspects of the question or task? 3. **Clarity**: Is the response clear and easy to understand? 4. **Conciseness**: Is the response appropriately concise without sacrificing important details? 5. **Relevance**: Does the response stay focused on the question or task at hand? Provide thoughtful reflections on these aspects and any other relevant factors. Use the score to indicate the overall quality, and set found_solution to true only if the response fully addresses the question or completes the task. ### Example Usage ``` reflections: "The response was clear and concise, addressing the main question effectively. However, it could have provided more context on edge cases." score: 8 found_solution: true ``` ## Question-Answering Trajectory Analyzer Guidelines for analyzing solution paths to question-answering tasks. ### Trajectory Components **Observations**: Environmental information about the current situation that provides context for decision-making. **Thoughts**: Reasoning about the current situation, analyzing what has been observed and planning next steps. **Actions**: The steps taken to progress toward solving the task. ### Action Types **Search[entity]**: Searches for the exact entity and returns relevant information if the entity exists. If not, returns suggestions for similar entities. **Lookup[keyword]**: Returns the next relevant passage that contains the keyword. Used for finding specific information within retrieved content. **Finish[answer]**: Returns the answer and finishes the task. Used when sufficient information has been gathered to provide a definitive response. ### Analysis Guidelines When analyzing a trajectory: 1. Evaluate whether each observation provides useful information 2. Assess if thoughts demonstrate logical reasoning 3. Determine if actions are appropriate given the current state 4. Score the trajectory correctness from 1-10 5. Evaluate reasoning validity even in incomplete trajectories 6. Do not generate additional steps; only analyze what is provided ## Prompt Engineering Patterns ### Chain of Thought Guide the model through step-by-step reasoning: ``` Let's approach this step by step: 1. First, identify the key components 2. Then, analyze each component 3. Finally, synthesize the findings ``` ### Few-Shot Learning Provide examples to establish the pattern: ``` Example 1: [input] -> [output] Example 2: [input] -> [output] Now apply this pattern to: [new input] ``` ### Self-Consistency Generate multiple reasoning paths and select the most consistent answer. ### Reflection Prompts Encourage self-critique: ``` Review your response and identify: - Any potential errors or oversights - Areas that could be explained more clearly - Missing information that would strengthen the answer ``` ## Quality Metrics ### Response Scoring Rubric - **10**: Perfect response, fully addresses all aspects with exceptional clarity - **8-9**: Excellent response with minor room for improvement - **6-7**: Good response that addresses the main points but lacks depth - **4-5**: Adequate response with significant gaps or unclear explanations - **2-3**: Poor response that misses key aspects or contains errors - **0-1**: Response fails to address the question or is completely incorrect ### Trajectory Scoring - **10**: Optimal path with efficient, logical steps - **7-9**: Good path with minor inefficiencies - **4-6**: Acceptable path but with unnecessary steps or missed opportunities - **1-3**: Poor path with fundamental reasoning errors
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.