Creating high-quality agent skills following Claude's official best practices. Use when designing, implementing, or improving agent skills, including naming conventions, progressive disclosure patterns, description writing, and appropriate freedom levels. Helps ensure skills are concise, well-structured, and optimized for context efficiency.
Provides best practices for creating efficient Claude skills. Use when designing, implementing, or improving agent skills to ensure proper naming, progressive disclosure, scope validation, and optimal context usage.
/plugin marketplace add taisukeoe/agentic-ai-skills-creator/plugin install skills-helper@agentic-skills-creatorThis skill inherits all available tools. When active, it can use any tool Claude has access to.
README.mdreferences/degrees-of-freedom.mdreferences/progressive-disclosure.mdreferences/workflows-and-validation.mdtests/scenarios.mdGuide for creating agent skills that follow Claude's official best practices.
Concise is Key: The context window is shared. Only add what Claude doesn't already know. Default assumption: Claude is already very smart.
Progressive Disclosure: Skills load in three levels:
Keep SKILL.md small: Target ~200 lines, maximum 500 lines. Move detailed content to reference files aggressively.
Single Responsibility: Each skill does one thing well.
Ask the user to clarify (adapt questions to context):
For clear requests (e.g., "format markdown tables"):
For ambiguous requests (e.g., "handle data", "help with files"):
Red flags requiring extra clarification:
Get clear sense of: purpose, usage examples, triggers, AND scope boundaries.
Before proceeding, verify the skill follows Single Responsibility:
Check naming:
handling-data - too broad, what data? what handling?data-helper - generic, unclear purposetransforming-csv-data - specific operation + data typevalidating-json-schemas - clear single purposeCheck boundaries:
If scope is too broad:
Do NOT proceed to file creation until scope is validated.
Freedom level controls how much latitude Claude has when following the skill:
Unless clearly LOW freedom: Read references/degrees-of-freedom.md and apply the decision framework. Cite the factors (fragility, context-dependency, consistency, error impact) in your justification.
If uncertain after reading the reference, ask the user.
Identify what to include:
Scripts (scripts/): Executable code for deterministic operations
References (references/): Documentation loaded as needed
Assets (assets/): Files used in output (templates, images, fonts)
skill-name/
├── SKILL.md (required - AI agent instructions)
├── README.md (optional - human-facing installation and usage guide)
├── references/ (optional)
├── tests/
│ └── scenarios.md (required - self-evaluation scenarios)
├── scripts/ (optional)
└── assets/ (optional)
README.md vs SKILL.md:
README.md, if present, should include:
tests/scenarios.md purpose)Avoid creating: INSTALLATION_GUIDE.md, CHANGELOG.md, or other redundant docs. Use README.md for human-facing documentation.
---
name: skill-name
description: What the skill does and when to use it
---
Naming (use gerund form):
processing-pdfs, analyzing-spreadsheets, managing-databaseshelper, utils, tools, anthropic-*, claude-*Description (third person, include WHAT and WHEN):
Be specific and include key terms. Description is the primary triggering mechanism.
Use imperative form. Keep small (target ~200 lines, max 500). Include only:
For complex multi-step processes, see references/workflows-and-validation.md.
Example pattern:
## Quick start
[Basic usage]
## Advanced features
**Feature A**: See [references/feature-a.md](references/feature-a.md)
**Feature B**: See [references/feature-b.md](references/feature-b.md)
Keep references one level deep. See references/progressive-disclosure.md for patterns.
For files >100 lines, include table of contents at top.
Organize by domain when appropriate:
skill/
├── SKILL.md
└── references/
├── domain_a.md
└── domain_b.md
Avoid: deeply nested references, duplicate information, generic file names.
Create tests/scenarios.md for self-evaluation with /evaluating-skills-with-models:
## Scenario: [Name]
**Difficulty:** Easy | Medium | Hard | Edge-case
**Query:** User request that triggers this skill
**Expected behaviors:**
1. [Action description]
- **Minimum:** What counts as "did it"
- **Quality criteria:** What "did it well" looks like
- **Haiku pitfall:** Common failure mode for smaller models
- **Weight:** 1-5 (importance)
**Output validation:** (optional)
- Pattern: `regex`
- Line count: `< N`
Why this format: Binary pass/fail doesn't differentiate models. Quality-based scoring reveals capability differences.
After completing SKILL.md and references, identify which tools the skill uses:
Bash(git status:*), WebSearch, Skill(other-skill))allowed-tools field to frontmatter if needed---
name: skill-name
description: ...
allowed-tools: "Bash(git status:*) Bash(git diff:*) WebSearch"
---
This field is experimental but helps agents pre-approve tool access.
Important considerations:
Read, Glob are already allowed by default - do not includeEdit, Write are destructive - do not pre-approveBash(git status:*) Bash(git diff:*) Bash(git log:*)Bash(git:*) (too broad, includes destructive operations like git push --force)❌ Windows-style paths (scripts\file.py)
❌ Too many options without a default
❌ Time-sensitive information
❌ Inconsistent terminology
❌ Deeply nested references
❌ Vague instructions
Progressive Disclosure: references/progressive-disclosure.md - Detailed patterns and examples
Degrees of Freedom: references/degrees-of-freedom.md - Guidance on appropriate freedom levels
Workflows and Validation: references/workflows-and-validation.md - Creating workflows with validation and feedback loops
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.