agent-specialization
Guide creation of focused single-purpose agents following the One Agent One Prompt One Purpose principle. Use when designing new agents, refactoring general agents into specialists, or optimizing agent context for a single task.
What this skill does
# Agent Specialization Skill Guide for creating focused, single-purpose agents that maximize effectiveness. ## When to Use - Designing new agents for workflows - Refactoring god-mode agents into specialists - Optimizing agent context usage - Creating eval-friendly agent architecture ## Core Principle > "One Agent, One Prompt, One Purpose" Every agent should: - Have exactly one purpose - Run exactly one prompt - Use the full context window for that purpose - Be reproducible and improvable ## Design Workflow ### Step 1: Identify the Single Purpose Ask: "What is the ONE question this agent answers?" | Good Purpose | Bad Purpose | | --- | --- | | "Classify this issue" | "Classify, plan, and implement" | | "Generate a patch plan" | "Fix all the bugs" | | "Review against spec" | "Review, test, and document" | ### Step 2: Determine Minimum Required Context Apply the Minimum Context Principle: ```markdown ## Required Context - [Specific file or section needed] - [Pattern or example needed] ## NOT Needed - [Documentation that's irrelevant] - [Code that won't be touched] ``` ### Step 3: Select Appropriate Tools Only include tools the agent will actually use: | Purpose | Tools | | --- | --- | | Classification | Read | | Planning | Read, Write, Glob | | Implementation | Read, Write, Edit, Bash | | Review | Read, Bash, Glob | | Documentation | Read, Write | ### Step 4: Choose Model Match model to task complexity: | Model | Best For | | --- | --- | | haiku | Classification, simple extraction | | sonnet | Planning, moderate reasoning | | opus | Complex implementation, critical decisions | ### Step 5: Design Focused Output Format Output should be: - Structured (JSON when appropriate) - Minimal (only what downstream needs) - Parseable (for automation) ## Agent Template ```markdown --- description: [Single sentence describing the ONE purpose] tools: [Only tools actually needed] model: [haiku/sonnet/opus based on complexity] --- # [Agent Name] You are a [role] agent. Your ONE purpose is to [specific task]. ## Your Capabilities - **[Tool]**: [How it supports the purpose] ## Process [Focused steps for the single purpose] ## Output Format [Structured output format] ## Rules 1. [Constraint that maintains focus] 2. [Another constraint] ``` ## Anti-Patterns to Avoid ### God Mode Agent ```markdown # BAD: Does everything You are an all-purpose assistant. Plan features, implement code, write tests, review changes, and create documentation. Handle any request. ``` ### Unfocused Output ```markdown # BAD: Returns everything Return a detailed analysis including history, context, alternatives, implications, and recommendations for all stakeholders. ``` ### Kitchen Sink Tools ```markdown # BAD: All tools enabled tools: [Read, Write, Edit, Bash, Glob, Grep, WebFetch, Task, ...] ``` ## Benefits of Specialization 1. **Full Context Window**: 100% for the task 2. **No Context Confusion**: Single objective 3. **Reproducible**: Same prompt, same behavior 4. **Improvable**: Can optimize independently 5. **Eval-Friendly**: Can A/B test models 6. **Debuggable**: Clear scope of responsibility ## Example: Specialized vs God Mode ### God Mode (Bad) ```markdown Handle the GitHub issue: 1. Classify it 2. Create a branch 3. Plan the implementation 4. Implement the feature 5. Write tests 6. Run tests 7. Review the implementation 8. Fix any issues 9. Create documentation 10. Create a PR ``` ### Specialized (Good) ```text /classify-issue → Issue Classifier Agent /generate-branch-name → Branch Namer Agent /feature → Plan Generator Agent /implement → Plan Implementer Agent /test → Test Runner Agent /review → Spec Reviewer Agent /patch → Patch Planner Agent /document → Documentation Generator Agent /pull-request → PR Creator Agent ``` Each agent does ONE thing well. ## Memory References - @one-agent-one-purpose.md - Full principle documentation - @minimum-context-principle.md - Context engineering guidance - @review-vs-test.md - Example of different purposes ## Version History - **v1.0.0** (2025-12-26): Initial release --- ## Last Updated **Date:** 2025-12-26 **Model:** claude-opus-4-5-20251101
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.