New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Skillful Subagent Creator

by curiositech in curiositech/some_claude_skills

Creates Claude subagents equipped with curated skills for solving problems within DAG workflows.

MITAuto-check passedAgent Workflows

Install Skillful Subagent Creator

skills CLI
$ npx skills add curiositech/some_claude_skills --skill skillful-subagent-creator -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install curiositech/some_claude_skills skillful-subagent-creator --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/curiositech/some_claude_skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/skillful-subagent-creator .claude/skills/skillful-subagent-creator && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
skillful-subagent-creator
GitHub stars
243
Token cost
~3k tokens
SKILL.md length
745 words
Files
3
Skills in repo
109
Repo updated
First seen
Licence
MIT

At a glance

Creates Claude subagents equipped with curated skills for solving problems within DAG workflows.

  • Works in 6 steps: Define the Role → Select Skills (2-5) → Write the 4-Section Prompt → …
  • Designing a specialist subagent
  • SKILL.md covers When to Use, Subagent Creation Process, Step 1: Define the Role and Step 2: Select Skills (2-5), plus 6 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Skillful Subagent Creator is an agent skill from curiositech/some_claude_skills. Creates Claude subagents equipped with curated skills for solving problems within DAG workflows. Use when designing a specialist subagent, selecting skills for a subagent, writing the 4-section subagent prompt, or wiring subagents into orchestration DAGs. Activate on "create subagent", "subagent with skills", "specialist agent", "agent DAG", "orchestrate agents", "skill-equipped agent". NOT for creating skills themselves (use skill-architect), general Claude Code usage, or single-agent prompting without skills.

Its SKILL.md is about 3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files (for example `.claude-plugin/plugin.json` and `CHANGELOG.md`).

It sits in Agent Workflows, covering Subagents. It works with TypeScript. The repository describes itself as: Claude skills that make my life easier. The licence is MIT.

When your agent uses it

  • Designing a specialist subagent
  • Selecting skills for a subagent
  • Writing the 4-section subagent prompt
  • Wiring subagents into orchestration DAGs

Example prompts

  • “create subagent”
  • “subagent with skills”
  • “specialist agent”
  • “/skillful-subagent-creator”

Requirements

  • Pre-approved tools (allowed-tools): Read, Write, Edit, Grep, Glob

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Define the Role
  2. Select Skills (2-5)
  3. Write the 4-Section Prompt
  4. Define Input/Output Contracts
  5. Wire into DAG Workflows
  6. Test

What it can do on your machine

Read from SKILL.md and the folder at commit 6713fc7. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Write
    • Edit
    • Grep
    • Glob

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are mermaid, markdown, json and yaml).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Skillful Subagent Creator loads about 3k tokens when it runs. Until then it costs about 136 tokens; SKILL.md has 745 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~136
When it runs · the whole SKILL.md, loaded when a task matches
~3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from curiositech/some_claude_skills at commit 6713fc7, republished under its MIT licence (© curiositech). 745 words, ~2,958 tokens.

Download SKILL.mdSave it as .claude/skills/skillful-subagent-creator/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
skillful-subagent-creator
description
Creates Claude subagents equipped with curated skills for solving problems within DAG workflows. Use when designing a specialist subagent, selecting skills for a subagent, writing the 4-section subagent prompt, or wiring subagents into orchestration DAGs. Activate on "create subagent", "subagent with skills", "specialist agent", "agent DAG", "orchestrate agents", "skill-equipped agent". NOT for creating skills themselves (use skill-architect), general Claude Code usage, or single-agent prompting without skills.
allowed-tools
Read, Write, Edit, Grep, Glob
argument-hint
[role-name] [domain] [skills-to-equip]
metadata.category
Productivity & Meta
metadata.tags
skillful, subagent, creator, create-subagent, subagent-with-skills

Skillful Subagent Creator

Creates Claude subagents that are equipped with curated skills as their standard operating procedures. Each subagent is "a specialist with a toolkit" — a narrow role, a small skill set, and a clear workflow for applying those skills.

When to Use

✅ Use for:

  • Designing a specialist subagent for a specific domain
  • Selecting which skills to attach to a subagent (2-5 core skills)
  • Writing the 4-section subagent prompt (Identity, Skills, Task Loop, Constraints)
  • Wiring subagents into DAG orchestration workflows
  • Defining input/output contracts between DAG nodes

❌ NOT for:

  • Creating the skills themselves (use skill-architect)
  • Single-agent prompting without skills
  • General Claude Code features or MCP setup

Subagent Creation Process

mermaid
flowchart TD
  A[Define the role] --> B[Select 2-5 skills]
  B --> C[Write 4-section prompt]
  C --> D[Define input/output contracts]
  D --> E[Choose DAG position]
  E --> F{Standalone or chained?}
  F -->|Standalone| G[Test with sample task]
  F -->|Chained| H[Define handoff protocol]
  H --> G
  G --> I{Works correctly?}
  I -->|No| C
  I -->|Yes| J[Deploy]

Step 1: Define the Role

A subagent's role must be narrow enough to be expert, broad enough to be useful.

mermaid
flowchart TD
  A{How narrow?} -->|Too broad| B["Do all marketing" ❌]
  A -->|Right| C["Draft landing page copy" ✅]
  A -->|Too narrow| D["Fix typos in H1 tags" ❌]

Test: Can you describe what this subagent does in one sentence with a specific verb and noun? If not, narrow it.

Examples of well-scoped roles:

RoleDomainSentence
RefactorerTypeScript"Designs and executes safe refactors in TypeScript monorepos"
PR ReviewerCode review"Reviews pull request diffs for correctness, style, and security"
Research SynthesizerResearch"Synthesizes multi-source research into cited briefing documents"
Migration PlannerDatabases"Plans database schema migrations with rollback strategies"

Step 2: Select Skills (2-5)

Skills are the subagent's standard operating procedures. Too few and it improvises; too many and it drowns in context.

Selection Criteria
mermaid
flowchart TD
  A{For each candidate skill} --> B{Needed for >50% of tasks?}
  B -->|Yes| C[Preload into context]
  B -->|No| D{Needed for 10-50%?}
  D -->|Yes| E[Add to dynamic catalog]
  D -->|No| F[Don't include]
Three Loading Tiers
TierSkillsHow LoadedToken Cost
Preloaded2-5 core skillsFull SKILL.md injected into system prompt~2-5k tokens per skill
Catalog5-20 available skillsName + 1-line description in prompt; loaded on demand~50 tokens per skill
NoneEverything elseNot mentioned; agent uses generic reasoning0 tokens
Example Skill Selection

Role: PR Reviewer for TypeScript/React

TierSkillWhy
Preloadedcode-review-skillCore to every task
Preloadedreact-server-componentsCatches RSC anti-patterns
Catalogtypescript-strict-modeSometimes relevant
Catalogtesting-patternsOnly for test file reviews
Nonedatabase-migrationNot in scope

Step 3: Write the 4-Section Prompt

Every subagent prompt has exactly four sections. This structure ensures the subagent knows its role, its tools, its workflow, and its constraints.

Section 1: Identity
markdown
You are the **[Role Name]** subagent for this system.
You handle [narrow domain of tasks].
When a task is outside this scope, explicitly say so and
ask the orchestrator for a different agent.

Keep it to 2-4 sentences. Name the specific domain. State the boundary explicitly.

Section 2: Skill Usage Rules
markdown
You have access to the following skills, which define your methods:
- `skill-a`: [1-line purpose]
- `skill-b`: [1-line purpose]
- `skill-c`: [1-line purpose]

These are your standard operating procedures, not optional hints.
When tackling a task:
1. Decide which skill(s) apply
2. Follow their step-by-step workflow
3. Use their output formats and checklists
4. Reference skill steps by number as you work
Section 3: Task-Handling Loop
markdown
For each task you receive:
1. Restate the task in your own words
2. Select one or more skills that fit. If none fit well, say so.
3. If needed, ask 2-5 clarifying questions
4. Produce a short internal plan
5. Execute the skill workflow step by step
6. Run any validation/QA steps from the skill
7. Return:
   (a) Final artifacts
   (b) Which skills you used (and which steps)
   (c) Assumptions and remaining risks
Section 4: Constraints
markdown
Quality bar: [e.g., "Never knowingly leave tests failing"]
Safety: [e.g., "No destructive operations without confirmation"]
Tie-breaking: [e.g., "If speed vs robustness conflict, pick robustness"]
Output format: [e.g., "Always return valid JSON matching the output contract"]

Step 4: Define Input/Output Contracts

Every subagent must have explicit contracts so orchestrators and downstream agents can work with it.

Input Contract
json
{
  "task": "string — description of what to do",
  "files": ["string — paths to relevant files"],
  "context": "string — prior agent output or user requirements",
  "constraints": {
    "time_budget": "string — e.g., '5 minutes'",
    "quality_bar": "string — e.g., 'production-ready'"
  }
}
Output Contract
json
{
  "status": "pass | warn | fail",
  "artifacts": ["string — files created or modified"],
  "summary": "string — 1-3 sentence description of what was done",
  "skills_used": ["string — skill names and step numbers"],
  "risks": ["string — remaining risks or assumptions"],
  "metadata": {
    "duration_ms": "number",
    "tokens_used": "number"
  }
}

Step 5: Wire into DAG Workflows

Orchestration Patterns
mermaid
flowchart TD
  subgraph "Single Specialist"
    O1[Orchestrator] --> S1[Specialist]
    S1 --> O1
  end

  subgraph "Chain"
    O2[Orchestrator] --> C1[Designer]
    C1 --> C2[Implementer]
    C2 --> C3[Tester]
    C3 --> O2
  end

  subgraph "Fan-out / Fan-in"
    O3[Orchestrator] --> P1[Auth Agent]
    O3 --> P2[Billing Agent]
    O3 --> P3[UI Agent]
    P1 --> M[Merger]
    P2 --> M
    P3 --> M
    M --> O3
  end
Pattern Selection
PatternWhen to UseSkill Requirement
Single SpecialistTask maps to one domainAgent has 2-5 preloaded skills
ChainSequential transformation pipelineEach agent has different skills; output of A feeds input of B
Fan-out / Fan-inIndependent parallel workAgents work concurrently; merger resolves conflicts
LoopIterative refinementSame agent re-runs with feedback until quality bar met
Human-in-the-LoopApproval requiredAgent produces draft; human reviews; agent revises
Show full SKILL.md (275 more words)Show less
DAG Node Definition

Each node in a DAG is a subagent with:

yaml
node:
  id: review-pr
  agent: pr-reviewer
  skills: [code-review-skill, react-server-components]
  input_from: [parse-diff]
  output_to: [merge-decision]
  retry: 2
  timeout: 300s
  on_failure: escalate-to-human

Step 6: Test

Test Protocol
  1. Happy path: Give the subagent a task that matches its skills perfectly
  2. Edge case: Give a task at the boundary of its scope
  3. Out of scope: Give a task outside its domain — it should refuse and say so
  4. Skill coverage: Verify it uses the attached skills, not ad-hoc reasoning
  5. Output contract: Verify output matches the JSON schema exactly
Red Flags
  • Agent invents processes instead of following skills → Skill usage rules not strong enough
  • Agent tries to handle out-of-scope tasks → Identity section not clear enough
  • Output doesn't match contract → Add explicit output format to constraints
  • Agent loads all references eagerly → Add lazy-loading instruction to prompt

Complete Example: PR Reviewer Subagent

Config
yaml
name: pr-reviewer
role: "Reviews TypeScript/React pull requests for correctness, style, and security"
skills:
  preloaded:
    - code-review-skill
    - react-server-components
  catalog:
    - typescript-strict-mode
    - testing-patterns
tools: [Read, Grep, Glob]
input_from: [diff-parser]
output_to: [merge-decision]
Prompt
You are the **PR Reviewer** subagent. You review TypeScript and React
pull request diffs for correctness, readability, performance, and security.
You do NOT implement features, fix bugs, or write new code. If asked to do
so, say "This is outside my scope — route to the Implementer agent."

Your skills define your standard process:
- `code-review-skill`: Structured review methodology (correctness → style → perf → security)
- `react-server-components`: Catches RSC anti-patterns ('use client' overuse, async component errors)

For each PR you receive:
1. Restate what the PR changes and why
2. Select applicable skills (always code-review-skill; add react-server-components if React files changed)
3. Follow the skill's review steps in order
4. Produce findings in the output contract format
5. Summarize: approve, request changes, or block

Quality bar: Never approve a PR with failing tests or security vulnerabilities.
Tie-breaking: If unsure whether something is a bug or style preference, flag it as "suggestion" not "required."
Output: Always return JSON matching the output contract schema.

Anti-Patterns

Skill-Less Subagent

Wrong: Creating a subagent with just a role description and no skills. Why: It will improvise a new process every time instead of following a consistent methodology. Fix: Attach 2-5 skills as standard operating procedures.

Skill Overload

Wrong: Attaching 15 skills to one subagent. Why: Blows context window; agent spends tokens selecting skills instead of executing. Fix: Max 5 preloaded, rest in catalog. Let the orchestrator pre-filter.

Missing Output Contract

Wrong: Subagent returns free-form text that downstream agents can't parse. Why: DAG breaks because the next node can't consume the output. Fix: Define explicit JSON output contract. Validate in tests.

Scope Creep

Wrong: Subagent tries to handle everything instead of refusing out-of-scope tasks. Why: Identity section doesn't state boundaries clearly enough. Fix: Add "You do NOT handle X, Y, Z. If asked, say so and suggest the right agent."

© curiositech, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files in .claude/skills/skillful-subagent-creator of curiositech/some_claude_skills.

  • SKILL.md
  • .claude-plugin/plugin.json
  • CHANGELOG.md

Open the folder on GitHubat commit 6713fc7

Compare with similar skills

Skillful Subagent Creator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Skillful Subagent Creator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Skillful Subagent Creator this skillcuriositech/some_claude_skills243—~3kAutomated safety check: PassMIT
Migrating Mastra To Pydantic AIpydantic/pydantic-ai20k—~2kAutomated safety check: PassMIT
Migrating Vercel AI SDK And Eve To Pydantic AIpydantic/pydantic-ai20k—~2.3kAutomated safety check: PassMIT
Deep Agents Orchestrationlangchain-ai/langchain-skills1.3k—~3.4kAutomated safety check: PassMIT
Golem Parallel Workers TSgolemcloud/golem1.5k—~2.2kAutomated safety check: PassCustom licence
Jevgrepdzhng/jevgrep2.4k—~741Automated safety check: PassMIT

Similar skills

  • Migrating Mastra To Pydantic AI

    pydantic/pydantic-ai

    Official

    Migrate TypeScript Mastra applications to Python with Pydantic AI and, only when needed, Pydantic AI Harness.

    20k GitHub stars~2k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Migrating Vercel AI SDK And Eve To Pydantic AI

    pydantic/pydantic-ai

    Official

    Migrate TypeScript Vercel AI SDK or Eve applications to Python with Pydantic AI and, only when needed, Pydantic AI Harness.

    20k GitHub stars~2.3k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Deep Agents Orchestration

    langchain-ai/langchain-skills

    Official

    INVOKE THIS SKILL when using subagents, task planning, or human approval in Deep Agents.

    1.3k GitHub stars~3.4k tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed

  • Golem Parallel Workers TS

    golemcloud/golem

    Fan out work to multiple parallel agents and collect results in a TypeScript Golem project.

    1.5k GitHub stars~2.2k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Jevgrep

    dzhng/jevgrep

    A skill your agent uses for questions about how, why, or where behavior works in a repository, including questions that name a function or setting.

    2.4k GitHub stars~741 tokensUpdated 6 days ago
    DevelopmentAuto-check passed

  • Deep Agents Core

    langchain-ai/langchain-skills

    Official

    Explains how to build agents with the Deep Agents framework: create_deep_agent, the built-in middleware, the harness, SKILL.md format and configuration options.

    1.3k GitHub starsUsed in 1 repo~3.1k tokens
    AI & LLM EngineeringAuto-check passed

More from curiositech/some_claude_skills

All 109 skills in this repo

  • Crisis Detection Intervention AI

    curiositech/some_claude_skills

    Detect crisis signals in user content using NLP, mental health sentiment analysis, and safe intervention protocols.

    243 GitHub starsUsed in 3 repos~3.8k tokens
    Auto-check passed

  • Form Validation Architect

    curiositech/some_claude_skills

    End-to-end form handling with react-hook-form, Zod schemas, validation patterns, error messaging, field arrays, and multi-step wizards.

    243 GitHub stars~3.8k tokensUpdated 1 mo ago
    Auto-check passed

  • Competitive Cartographer

    curiositech/some_claude_skills

    Strategic analyst that maps competitive landscapes, identifies white space opportunities, and provides positioning recommendations.

    243 GitHub starsUsed in 1 repo~1.3k tokens
    Auto-check passed

  • GitHub Actions Pipeline Builder

    curiositech/some_claude_skills

    Build production CI/CD pipelines with GitHub Actions. An agent skill from curiositech/some_claude_skills.

    243 GitHub stars~2.8k tokensUpdated 1 mo ago
    Auto-check: notes

  • Computer Vision Pipeline

    curiositech/some_claude_skills

    Build production computer vision pipelines for object detection, tracking, and video analysis.

    243 GitHub starsUsed in 1 repo~4k tokens
    Auto-check passed

  • Design Archivist

    curiositech/some_claude_skills

    Long-running design anthropologist that builds comprehensive visual databases from 500-1000 real-world examples, extracting color palettes, typography patterns, layout systems, and interaction…

    243 GitHub starsUsed in 1 repo~1.6k tokens
    Auto-check passed

Works with

Categories

Questions about Skillful Subagent Creator

What does Skillful Subagent Creator do?

Creates Claude subagents equipped with curated skills for solving problems within DAG workflows. Skillful Subagent Creator is an agent skill from curiositech/some_claude_skills. Creates Claude subagents equipped with curated skills for solving problems within DAG workflows.

When should I use Skillful Subagent Creator?

Skillful Subagent Creator fits situations like: designing a specialist subagent; selecting skills for a subagent; writing the 4-section subagent prompt; wiring subagents into orchestration DAGs.

How do I install Skillful Subagent Creator in Claude Code?

Run `npx skills add curiositech/some_claude_skills --skill skillful-subagent-creator -a claude-code`. Or copy the skill folder (.claude/skills/skillful-subagent-creator in curiositech/some_claude_skills) into .claude/skills/skillful-subagent-creator in your project. Claude Code loads it when a task matches its description.

How do I install Skillful Subagent Creator in Codex?

Run `npx skills add curiositech/some_claude_skills --skill skillful-subagent-creator -a codex`. Or copy the skill folder (.claude/skills/skillful-subagent-creator in curiositech/some_claude_skills) into .agents/skills/skillful-subagent-creator in your project. Codex loads it when a task matches its description.

Can I use Skillful Subagent Creator in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add curiositech/some_claude_skills --skill skillful-subagent-creator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skillful-subagent-creator, .gemini/skills/skillful-subagent-creator, .github/skills/skillful-subagent-creator and .opencode/skills/skillful-subagent-creator in your project.

What does Skillful Subagent Creator need to run?

SKILL.md names no scripts, command-line tools or credentials: Skillful Subagent Creator is instructions for the agent only. Its frontmatter pre-approves these tools: Read, Write, Edit, Grep, Glob.

Does Skillful Subagent Creator access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Skillful Subagent Creator safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Skillful Subagent Creator use?

Skillful Subagent Creator is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Skillful Subagent Creator use?

About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Skillful Subagent Creator?

Skills that share tags, products or a category with Skillful Subagent Creator: Migrating Mastra To Pydantic AI (pydantic/pydantic-ai, 20k stars), Migrating Vercel AI SDK And Eve To Pydantic AI (pydantic/pydantic-ai, 20k stars), Deep Agents Orchestration (langchain-ai/langchain-skills, 1.3k stars) and Golem Parallel Workers TS (golemcloud/golem, 1.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Skillful Subagent Creator?

curiositech (a GitHub organization) maintains it in curiositech/some_claude_skills, which has 243 GitHub stars. The repository holds 109 skills in this directory. The repository was last updated on September 6, 2026.

Source: curiositech/some_claude_skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.