Agent skill

Nav Skill Creator

by qf-studio in qf-studio/navigator

Analyze codebase patterns and create custom skills for repetitive workflows.

MITAuto-check: notesAgent Workflows

Install Nav Skill Creator

skills CLI
$ npx skills add qf-studio/navigator --skill nav-skill-creator -a claude-code

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

GitHub CLI
$ gh skill install qf-studio/navigator nav-skill-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/qf-studio/navigator.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/nav-skill-creator .claude/skills/nav-skill-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
nav-skill-creator
GitHub stars
355
Token cost
~3.1k tokens
SKILL.md length
787 words
Files
4
Skills in repo
32
Repo updated
First seen
Licence
MIT

At a glance

Analyze codebase patterns and create custom skills for repetitive workflows.

  • Works in 6 steps: Understand Skill Request → Analyze Codebase Patterns → Design Skill Structure → …
  • Project needs automation
  • SKILL.md covers When to Invoke, What This Does, Execution Steps and Example Workflows, plus 4 more sections
  • Runs Python scripts from its folder

What it does

Nav Skill Creator is an agent skill from qf-studio/navigator. Analyze codebase patterns and create custom skills for repetitive workflows. Use when project needs automation or pattern enforcement. Auto-invoke when user says "create a skill for...", "automate this workflow", or "we keep doing X manually".

Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 6 other files (for example `examples/example-generated-skill.md`, `functions/skill_generator.py` and `templates/skill-template.md`).

It sits in Agent Workflows, covering Skill authoring. The repository describes itself as: Finish What You Start — Context engineering for Claude Code. Sessions last 20+ exchanges instead of crashing at 7. The licence is MIT.

When your agent uses it

  • Project needs automation
  • Pattern enforcement
  • Says create a skill for...
  • Automate this workflow

Example prompts

  • “create a skill for...”
  • “automate this workflow”
  • “we keep doing X manually”
  • “/nav-skill-creator”

Requirements

  • Python 3
  • Docker
  • Pre-approved tools (allowed-tools): Read, Write, Edit, Grep, Glob, Bash, Task

Workflow steps

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

  1. Understand Skill Request
  2. Analyze Codebase Patterns
  3. Design Skill Structure
  4. Generate Skill Files
  5. Test Generated Skill
  6. Document New Skill

What it can do on your machine

Read from SKILL.md and the folder at commit 3bb9eac. 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
    • Bash
    • Task

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships script files (Python), which the agent can run.

    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

Nav Skill Creator loads about 3.1k tokens when it runs. Until then it costs about 65 tokens; SKILL.md has 787 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~65
When it runs · the whole SKILL.md, loaded when a task matches
~3.1k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Write, Edit, Grep, Glob, Bash, Task

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 qf-studio/navigator at commit 3bb9eac, republished under its MIT licence (© qf-studio). 787 words, ~3,095 tokens.

Download SKILL.mdSave it as .claude/skills/nav-skill-creator/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
nav-skill-creator
description
Analyze codebase patterns and create custom skills for repetitive workflows. Use when project needs automation or pattern enforcement. Auto-invoke when user says "create a skill for...", "automate this workflow", or "we keep doing X manually".
allowed-tools
Read, Write, Edit, Grep, Glob, Bash, Task
version
1.0.0

Navigator Skill Creator

Create project-specific skills by analyzing codebase patterns and automating repetitive workflows.

When to Invoke

Auto-invoke when user mentions:

  • "Create a skill for [pattern]"
  • "Automate this workflow"
  • "We keep doing X manually"
  • "Enforce this pattern"
  • "Generate boilerplate for [feature type]"
  • "We need consistency for [task type]"

What This Does

  1. Analyzes codebase to understand project patterns
  2. Identifies best practices from existing code
  3. Generates skill with:
    • Auto-invocation triggers
    • Predefined functions
    • Templates
    • Examples
  4. Tests the generated skill
  5. Documents the new skill

Execution Steps

Step 1: Understand Skill Request

Ask clarifying questions:

  • What pattern/workflow to automate?
  • What triggers should invoke this skill?
  • What output format is expected?
  • Are there existing examples in the codebase?

Example dialogue:

User: "Create a skill for adding React components"
Assistant: "I'll analyze your codebase to understand React component patterns.
- What directory are components in?
- Do you use TypeScript or JavaScript?
- Do you want tests generated automatically?
- Are there style files (CSS/SCSS) per component?"
Step 2: Analyze Codebase Patterns

Use Task agent to explore (saves 60-80% tokens):

Use Task agent with subagent_type=Explore:
"Find existing [pattern type] in codebase:
 - Locate all [files matching pattern]
 - Identify common structure
 - Extract best practices
 - Find configuration files
 - Return summary of findings"

What to look for:

  • File naming conventions (kebab-case, PascalCase, etc.)
  • Directory structure patterns
  • Import/export patterns
  • Testing patterns
  • Configuration patterns
  • Documentation patterns

Example for React components:

Task agent finds:
- Components in src/components/
- PascalCase naming (UserProfile.tsx)
- Co-located tests (UserProfile.test.tsx)
- Props interfaces defined above component
- Export default at bottom
Step 3: Design Skill Structure

Determine skill metadata:

yaml
name: [project]-[pattern-type]
description: [When to auto-invoke + what it does]
allowed-tools: [Read, Write, Edit, Grep, Glob, Bash, Task]
version: 1.0.0

Plan directory structure:

skills/[skill-name]/
├── SKILL.md              # Main instructions
├── functions/            # Python helper scripts
│   └── [generator].py
├── examples/             # Reference implementations
│   └── [example].[ext]
└── templates/            # Output format templates
    └── [template].[ext]

Design predefined functions:

  • What repetitive logic can be automated?
  • What validation should be enforced?
  • What formatting ensures consistency?

Example functions for frontend-component skill:

  • component_generator.py - Generate component boilerplate
  • test_generator.py - Generate test file
  • style_generator.py - Generate style file
  • name_validator.py - Validate component naming
Step 4: Generate Skill Files

4.1 Create SKILL.md

markdown
---
name: [skill-name]
description: [Auto-invocation triggers + purpose]
allowed-tools: [List of tools]
version: 1.0.0
---

# [Skill Title]

[Brief description of what this skill does]

## When to Invoke

Auto-invoke when user says:
- "[trigger phrase 1]"
- "[trigger phrase 2]"
- "[trigger phrase 3]"

## What This Does

1. [Step 1 overview]
2. [Step 2 overview]
3. [Step 3 overview]

## Execution Steps

### Step 1: [Step Name]

[Detailed instructions for this step]

**Use predefined function**: `functions/[function-name].py`

4.2 Create Predefined Functions

python
# functions/[generator].py

def generate_[output](name, config):
    """
    Generate [output type] based on project patterns.

    Args:
        name: [Description]
        config: [Description]

    Returns:
        [output]: [Description]
    """
    # Implementation based on codebase analysis
    pass

4.3 Create Examples

examples/
└── [reference-implementation].[ext]
    - Real example from codebase (best practice)
    - Shows expected structure
    - Demonstrates conventions

4.4 Create Templates

templates/
└── [output-template].[ext]
    - Skeleton structure with placeholders
    - ${VAR_NAME} for substitution
    - Comments explaining sections
Step 5: Test Generated Skill

5.1 Verify skill loads:

bash
# In project root
grep -r "name: [skill-name]" skills/

5.2 Test auto-invocation:

In Claude Code conversation:
"[Use one of the auto-invoke trigger phrases]"

Expected: Skill should be detected and loaded

5.3 Test execution:

  • Run through skill steps
  • Verify functions work correctly
  • Check output matches template
  • Validate generated code follows patterns

5.4 Iterate if needed:

  • Fix function bugs
  • Improve templates
  • Add missing examples
  • Clarify instructions
Step 6: Document New Skill

Update project documentation:

  1. CLAUDE.md - Add to skills section:
markdown
#### [Skill Name]
**Auto-invoke**: "[trigger phrase]"
**Purpose**: [What it does]
**Generates**: [Output type]
  1. README.md - Add to skills list:
markdown
- **[skill-name]**: [Brief description]
  1. .agent/system/plugin-patterns.md - Add to skill registry:
markdown
### [Skill Name]
**Created**: [Date]
**Pattern**: [What pattern it enforces]
**Functions**: [List of predefined functions]

Register in plugin.json (if applicable):

json
{
  "skills": [
    {
      "name": "[skill-name]",
      "path": "skills/[skill-name]/SKILL.md"
    }
  ]
}

Example Workflows

Example 1: Create Skill for Adding API Endpoints

User: "Create a skill for adding REST API endpoints"

Execution:

  1. Clarify:

    • Which framework? (Express, Fastify, etc.)
    • Where are routes defined?
    • Authentication required?
    • Testing strategy?
  2. Analyze (via Task agent):

    Find existing API endpoints:
    - Routes in api/routes/
    - Controllers in api/controllers/
    - Middleware in api/middleware/
    - Tests in tests/api/
  3. Design:

    yaml
    name: backend-api-endpoint
    description: Add new REST API endpoint following project conventions. Use when user says "add endpoint", "create API", or "new route".
    allowed-tools: Read, Write, Edit, Grep, Glob, Bash
  4. Generate:

    skills/backend-api-endpoint/
    ├── SKILL.md
    ├── functions/
    │   ├── endpoint_generator.py
    │   └── route_validator.py
    ├── examples/
    │   └── user-endpoint.ts
    └── templates/
        ├── route-template.ts
        └── test-template.spec.ts
  5. Test:

    User: "Add a POST /posts endpoint"
    Skill: Auto-invoked, generates route + controller + test
    Verify: Files follow project conventions
  6. Document: Update CLAUDE.md, README.md, plugin-patterns.md

Example 2: Create Skill for React Components

User: "Automate creating new React components"

Execution:

  1. Clarify:

    • TypeScript or JavaScript?
    • Functional or class components?
    • Style approach? (CSS modules, styled-components, etc.)
    • Test library? (Jest, React Testing Library, etc.)
  2. Analyze (via Task agent):

    Find React components:
    - Components in src/components/
    - PascalCase naming
    - TypeScript (.tsx)
    - CSS modules (.module.css)
    - Tests with RTL
  3. Design:

    yaml
    name: frontend-component
    description: Create new React component with TypeScript, styles, and tests. Use when user says "create component", "add component", or "new React component".
    allowed-tools: Read, Write, Edit, Grep, Glob, Bash
  4. Generate:

    skills/frontend-component/
    ├── SKILL.md
    ├── functions/
    │   ├── component_generator.py
    │   ├── test_generator.py
    │   └── style_generator.py
    ├── examples/
    │   ├── Button.tsx
    │   └── Button.test.tsx
    └── templates/
        ├── component-template.tsx
        ├── test-template.tsx
        └── style-template.module.css
  5. Test:

    User: "Create a UserProfile component"
    Skill: Auto-invoked, generates component + test + styles
    Verify: Props interface, exports, naming correct
  6. Document: Update project docs


Output Format

After generating skill, show summary:

✅ Skill Created: [skill-name]

Structure:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📁 skills/[skill-name]/
   ├── SKILL.md
   ├── functions/
   │   └── [N functions created]
   ├── examples/
   │   └── [N examples added]
   └── templates/
       └── [N templates created]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Auto-Invocation Triggers:
- "[trigger 1]"
- "[trigger 2]"
- "[trigger 3]"

Next Steps:
1. Test the skill: "[example trigger phrase]"
2. Iterate if needed
3. Documentation updated

Try it now: "[example usage]"

Show full SKILL.md (374 more words)Show less

Best Practices

Pattern Analysis
  • Use Task agent for codebase exploration (saves 60-80% tokens)
  • Look at 3-5 examples minimum (find patterns vs outliers)
  • Identify conventions explicitly followed
  • Note edge cases in comments
Skill Design
  • Keep skills focused (one pattern per skill)
  • Clear auto-invocation triggers (3-5 phrases)
  • Minimal tools needed (add only what's required)
  • Progressive disclosure (details in functions, not main instructions)
Function Creation
  • One function = one responsibility
  • Type hints and docstrings required
  • Handle errors gracefully
  • Return structured data (not print statements)
Template Design
  • Use clear placeholders (${VAR_NAME})
  • Include comments explaining sections
  • Follow project style guide
  • Provide sensible defaults
Testing
  • Test with real project context
  • Verify auto-invocation works
  • Check output against best practices
  • Iterate based on actual usage

Common Patterns to Automate

Backend Patterns
  • REST API endpoints
  • GraphQL resolvers
  • Database migrations
  • Background jobs
  • Middleware functions
  • Authentication guards
Frontend Patterns
  • React/Vue/Svelte components
  • Redux/Vuex store modules
  • API client functions
  • Form validation schemas
  • Route definitions
  • Style component creation
Infrastructure Patterns
  • Docker service configs
  • CI/CD pipeline steps
  • Deployment scripts
  • Environment configs
  • Monitoring setup
Documentation Patterns
  • API documentation
  • Component documentation
  • Architecture decision records (ADRs)
  • Runbook entries
  • Changelog entries

Troubleshooting

Skill Not Auto-Invoking

Problem: Skill created but doesn't trigger automatically

Solutions:

  1. Check description has clear trigger phrases
  2. Verify plugin.json includes skill registration
  3. Reload Claude Code to refresh skill index
  4. Test with exact trigger phrase from description
Functions Not Executing

Problem: Predefined functions throw errors

Solutions:

  1. Check Python syntax is valid
  2. Verify function imports are correct
  3. Test function independently first
  4. Check error messages in execution logs
Templates Not Matching Output

Problem: Generated code doesn't match project conventions

Solutions:

  1. Re-analyze codebase for missed patterns
  2. Update templates with correct structure
  3. Add more examples showing variations
  4. Validate against linter/formatter
Skill Too Broad

Problem: Skill tries to do too much

Solutions:

  1. Split into multiple focused skills
  2. Remove optional features to separate skills
  3. Keep core pattern simple
  4. Add extensions as separate skills

Success Criteria

This skill succeeds when:

  • New skill auto-invokes correctly
  • Generated output follows project conventions
  • Functions execute without errors
  • Templates produce valid code
  • Examples are clear and relevant
  • Documentation is updated
  • Skill saves time vs manual work

The skill-creator is Navigator's self-improving engine - it learns your patterns and automates them 🔄

© qf-studio, 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 3 other files in skills/nav-skill-creator of qf-studio/navigator.

  • SKILL.md
  • examples/example-generated-skill.md
  • functions/skill_generator.py
  • templates/skill-template.md

Open the folder on GitHubat commit 3bb9eac

Compare with similar skills

Nav Skill 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.

Nav Skill Creator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Nav Skill Creator this skillqf-studio/navigator355—~3.1kAutomated safety check: NotesMIT
Skill CreatorAzure/azqr79689 repos~8.2kAutomated safety check: PassApache-2.0
Claude Code Skill Developer Guidediet103/claude-code-infrastructure-showcase10k11 repos~3.5kAutomated safety check: PassMIT
Darwin Skill Optimizeralchaincyf/darwin-skill6.2k1 repos~4.7kAutomated safety check: PassMIT
Claude Code Command Developmentanthropics/claude-plugins-official38k10 repos~4.8kAutomated safety check: PassApache-2.0
Claude Code Plugin Structureanthropics/claude-plugins-official38k10 repos~3.4kAutomated safety check: PassApache-2.0

Similar skills

  • Skill Creator

    Azure/azqr

    Official

    Create new skills, modify and improve existing skills, and measure skill performance.

    796 GitHub starsUsed in 89 repos~8.2k tokens
    Agent WorkflowsAuto-check passed
  • Claude Code Skill Developer Guide

    diet103/claude-code-infrastructure-showcase

    A guide to creating and managing Claude Code skills with auto-activation: skill-rules.json triggers, hooks, enforcement levels, YAML frontmatter and progressive disclosure.

    10k GitHub starsUsed in 11 repos~3.5k tokens
    Agent WorkflowsAuto-check passed
  • Darwin Skill Optimizer

    alchaincyf/darwin-skill

    Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.

    6.2k GitHub starsUsed in 1 repo~4.7k tokens
    Agent WorkflowsAuto-check passed
  • Claude Code Command Development

    anthropics/claude-plugins-official

    Official

    Explains how to write Claude Code slash commands: Markdown files with YAML frontmatter, arguments, file references, bash context and interactive prompts.

    38k GitHub starsUsed in 10 repos~4.8k tokens
    Agent WorkflowsAuto-check passed
  • Claude Code Plugin Structure

    anthropics/claude-plugins-official

    Official

    Explains the directory layout, plugin.json manifest and component organization of a Claude Code plugin, including auto-discovery and portable paths.

    38k GitHub starsUsed in 10 repos~3.4k tokens
    Agent WorkflowsAuto-check passed
  • Skill Release Gate

    rohitg00/ai-engineering-from-scratch

    Evaluates an Agent Skill bundle before release for structure, trigger quality, artifact improvement, script correctness, safety, installed-tree integrity and host portability.

    67k GitHub stars~1k tokensUpdated today
    Agent WorkflowsAuto-check passed

More from qf-studio/navigator

All 32 skills in this repo
  • Nav Start

    qf-studio/navigator

    Load Navigator documentation navigator when starting development session, resuming work, or beginning new feature.

    355 GitHub stars~4.7k tokensUpdated 2 days ago
    Auto-check: notes
  • Backend Endpoint

    qf-studio/navigator

    Create REST/GraphQL API endpoint with validation, error handling, and tests.

    355 GitHub stars~4.5k tokensUpdated 2 days ago
    Auto-check: notes
  • Backend Test

    qf-studio/navigator

    Generate backend tests (unit, integration, mocks) for existing code.

    355 GitHub stars~1.5k tokensUpdated 2 days ago
    Auto-check: notes
  • Database Migration

    qf-studio/navigator

    Create database migration with schema changes and rollback. An agent skill from qf-studio/navigator.

    355 GitHub stars~3.7k tokensUpdated 2 days ago
    Auto-check: notes
  • Frontend Component

    qf-studio/navigator

    Create React/Vue component with TypeScript, tests, and styles.

    355 GitHub stars~4.5k tokensUpdated 2 days ago
    Auto-check: notes
  • Frontend Test

    qf-studio/navigator

    Generate frontend component tests (React Testing Library, Vue Test Utils, snapshot) for existing components.

    355 GitHub stars~1.6k tokensUpdated 2 days ago
    Auto-check: notes

Categories

Questions about Nav Skill Creator

What does Nav Skill Creator do?

Analyze codebase patterns and create custom skills for repetitive workflows. Nav Skill Creator is an agent skill from qf-studio/navigator. Analyze codebase patterns and create custom skills for repetitive workflows.

When should I use Nav Skill Creator?

Nav Skill Creator fits situations like: project needs automation; pattern enforcement; says create a skill for..; automate this workflow.

How do I install Nav Skill Creator in Claude Code?

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

How do I install Nav Skill Creator in Codex?

Run `npx skills add qf-studio/navigator --skill nav-skill-creator -a codex`. Or copy the skill folder (skills/nav-skill-creator in qf-studio/navigator) into .agents/skills/nav-skill-creator in your project. Codex loads it when a task matches its description.

Can I use Nav Skill 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 qf-studio/navigator --skill nav-skill-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/nav-skill-creator, .gemini/skills/nav-skill-creator, .github/skills/nav-skill-creator and .opencode/skills/nav-skill-creator in your project.

What does Nav Skill Creator need to run?

Going by SKILL.md and its folder, Nav Skill Creator needs Python for the scripts in its folder. Our summary lists: Python 3; Docker. Its frontmatter pre-approves these tools: Read, Write, Edit, Grep, Glob, Bash, Task.

Does Nav Skill 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 Nav Skill Creator safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Nav Skill Creator use?

Nav Skill 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 Nav Skill Creator use?

About 3.1k 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 Nav Skill Creator?

Skills that share tags, products or a category with Nav Skill Creator: Skill Creator (Azure/azqr, 796 stars), Claude Code Skill Developer Guide (diet103/claude-code-infrastructure-showcase, 10k stars), Darwin Skill Optimizer (alchaincyf/darwin-skill, 6.2k stars) and Claude Code Command Development (anthropics/claude-plugins-official, 38k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Nav Skill Creator?

qf-studio (a GitHub organization) maintains it in qf-studio/navigator, which has 355 GitHub stars. The repository holds 32 skills in this directory. The repository was last updated on October 8, 2026.

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