Built for AI-Enabled Engineering Teams

Every developer configured
their own way.
That ends today.

Your developers are using different AI tools, configured differently, with no shared standards and no visibility into what anyone else is doing. New hires start from scratch.  Configs drift.  Velocity suffers. prompticorn gives your whole team consistent, role-aware AI agents — one config system, committed to the repo, works for every tool.

Set Up Your Team See How It Works →
5
AI Tools Supported
1
Config Per Project
0
Config Drift
∞
Team Scale
Unified Prompt Architecture

One Prompt Architecture.
Five Tools.
Zero Compromise.

Run prompticorn init once. prompticorn generates production-grade configurations for Kilo, Cline, Claude, Copilot, and Cursor — automatically. Switch tools, swap personas, change variants. Your agents travel with you.

Deploy Now Architecture →
5
Tools
1×
Definition
0
Config Drift
∞
Scale
One config. Five AI tools. Zero drift.

Define once.
Deploy everywhere.

Stop maintaining five separate agent configs. Run prompticorn init once — prompticorn generates production-ready configurations for Kilo, Cline, Claude, Cursor, and Copilot automatically.

Get Started View on GitHub
⚡

5 AI Tools

Kilo, Cline, Claude, Cursor, Copilot

⚙️

Zero Config

Auto-discovery registry

🎯

Persona Filtering

Team-role aware generation

🗜️

Token Efficient

Minimal/verbose variants

Install: pip install prompticorn
The Problem

The AI tooling mess your team is living in

Whether your team has embraced AI agents or is still figuring it out, the same pattern emerges. Fragmentation, drift, and wasted hours nobody planned for.

Scenario A

Not using agents yet

Developers are using ChatGPT, Copilot autocomplete, or nothing structured. No shared AI workflow — each developer improvises. Onboarding new engineers means passing around .md files in Slack. Nobody knows what "the right way" to use AI in your codebase actually is. You're leaving serious velocity on the table.

Scenario B

Already using agents — but it's chaos

One dev uses Cursor, another uses Cline, another uses Claude Code. Each has their own agent config — written differently, named differently, maintained in isolation. When someone new joins, they start from scratch. Config drift is invisible until something breaks.

Without prompticorn
  • Each developer configures their own AI agents from scratch
  • No shared standards — "code review" means something different to everyone
  • New hires spend days figuring out how the team uses AI
  • Switching tools means rewriting every config
  • Frontend and backend devs get the same bloated agent set
  • Configs drift silently — nobody knows what's current
With prompticorn
  • One config file, shared across the whole team
  • Consistent agent behavior for everyone, every tool
  • New hires run one command and they're ready
  • Switch tools in seconds — configs regenerate automatically
  • Persona filtering: each role gets exactly what they need
  • prompticorn validate catches drift before it ships
Capabilities

Built for production teams

Every component engineered for serious engineering organizations. No toy features, no dead weight.

📐

Configure Once, Deploy Anywhere

Tool-agnostic configuration for Agent, Skill, Workflow, Tool, Rules, and Project. The single source of truth for your entire AI configuration estate.

🏗️

Builder Architecture

Five purpose-built builders transform your configuration into tool-specific outputs. BuilderFactory handles instantiation with a clean, extensible API surface.

🔍

Auto-Discovery Registry

Zero-config agent registration. The central Registry manages modes, files, and output ordering. Drop in an agent — it's automatically indexed.

👤

Persona-Based Filtering

Select your team's roles — Frontend, Backend, DevOps, QA/Tester, Data Scientist — and get only the relevant agents. Universal agents always included.

✂️

Minimal / Verbose Variants

Choose output verbosity at build time. Minimal variants cut token overhead. Verbose variants retain full context for complex workflows.

🧩

Jinja2 Template Engine

Custom filters, resolvers, and template handlers built on Jinja2. Full programmatic control over rendering logic with composable template components.

⌨️

CLI Toolchain

Six production commands: init, list, switch, swap, update, validate. Complete lifecycle management.

📦

Markdown / YAML Loader

Loader and Parser subsystems handle Markdown and YAML ingestion. Handles both legacy and current configuration formats without breakage.

🔄

Backwards Compatibility

Existing configurations continue to operate without modification. Adopt prompticorn incrementally — no forced rewrites, no big bang migrations.

Core Capabilities

Everything your team needs.
Nothing you don't.

Built for engineering teams that take AI tooling seriously. prompticorn eliminates configuration drift and boilerplate sprawl.

📐

Configure Once, Deploy Anywhere

A tool-agnostic configuration layer. Define Agent, Skill, Workflow, Tool, Rules, and Project models once — prompticorn handles translation to each tool's native format.

👤

Persona-Based Filtering

Select team roles — Frontend, Backend, DevOps, QA/Tester — and get only relevant agents. Universal agents (ask, debug, explain, plan, orchestrator) always included.

✂️

Minimal / Verbose Variants

Every builder supports output verbosity selection at build time. Choose minimal to save tokens, or verbose for rich, self-documenting configurations.

🔍

Auto-Discovery Registry

Zero-config agent registration. The BuilderFactory and Registry automatically discover and register builders. Edit .prompticorn.yaml directly or use prompticorn update.

🧩

Jinja2 Template Engine

A full Jinja2-powered rendering pipeline with custom filters, resolvers, and template handlers. Deterministic, testable, extensible output generation.

🔄

Backwards Compatible

Existing configurations continue to work. The Loader and Parser subsystems handle both legacy Markdown/YAML and the new config format.

⌨️

CLI First

Rich CLI covering full lifecycle: init, list, switch, swap, update, and validate. Integrate into onboarding scripts or CI pipelines.

✅

Validate Before You Ship

Built-in validation with prompticorn validate checks for correctness, missing references, and schema compliance before any output is generated.

How It Works

Three steps. Your whole team, aligned.

No build step. No CI pipeline changes. No DevOps work. Two minutes to set up. Done.

01

Initialize

Run prompticorn init in your project. Answer a few questions about your team's tools, roles, and preferences. Takes two minutes. prompticorn generates a .prompticorn.yaml config file.

02

Commit

Commit .prompticorn.yaml to your repository. Now every developer on your team — current and future — has access to the same agent configuration. One file. Source of truth.

03

Generate

Each developer runs prompticorn init locally. Their AI tool (Cursor, Cline, Claude Code, Copilot, Kilo) gets configured automatically — tailored to their role. Zero noise.

What prompticorn init asks
  1. Which AI tools does your team use?
  2. What is your primary programming language?
  3. Which team roles (personas) need to be supported?

Output: configs generated for every selected tool, personas defined, .prompticorn.yaml committed and ready to share.

Under the Hood

How it works technically

Optional reading — but it's here to satisfy the engineers on your team who want to verify the design is solid.

Component Graph

Interactive · D3.js

Build Pipeline

Mermaid
Architecture

Precision-engineered data flow

From a single definition to five tool-specific outputs. Every step deterministic, auditable, and reproducible.

01

Configure Your Agents

Run prompticorn init or edit .prompticorn.yaml to define your agents, skills, and workflows. YAML or Markdown — both supported. One definition, no duplication.

02

CLI Invocation

Run prompticorn init. The CLI queries the Registry to determine active mode, target tools, and persona configuration. No magic, no black boxes.

03

Persona Filtering

PersonaFilter applies team-role rules against your config. Irrelevant agents are excluded. Each persona receives a precisely scoped agent set — no noise, no bloat.

04

Builder Factory Dispatch

BuilderFactory instantiates the correct builder for each target tool. Kilo, Cline, Claude, Copilot, Cursor — each builder encapsulates its tool's specific output contract.

05

Template Engine Rendering

Jinja2 templates with custom filters and resolvers transform your config into final text. Minimal or verbose variant applied. Output is deterministic.

06

Files Written to Disk

Builders write to canonical locations: .kilo/agents/, .clinerules, .claude/, .github/instructions/, and .cursor/rules/.

Component Graph

Interactive · D3.js

Build Pipeline

Mermaid

Data Model (ERD)

Mermaid

CLI State Machine

Mermaid
Architecture

Under the hood.
Elegantly simple.

A clean pipeline from your configuration to tool-specific output. No magic — just a well-structured transformation layer.

01

Configure Your Agents

Run prompticorn init or edit .prompticorn.yaml to define your agents, skills, and workflows. YAML or Markdown — the loader handles both.

02

Registry Lookup

The CLI queries the central Registry for available modes, output ordering, and builder configuration. BuilderFactory instantiates the correct builder.

03

Persona Filter

If a persona is active, PersonaFilter selects only the agents relevant to that team role. Irrelevant agents are excluded from the build pass.

04

Template Render

Each Builder passes your config through Jinja2 template handlers. Custom filters and resolvers produce tool-specific syntax — YAML frontmatter, pure Markdown, or structured files.

05

Write to Disk

Builders write to canonical locations: .kilo/agents/, .clinerules, .claude/, .github/instructions/, .cursor/rules/.

Component Graph

Interactive · D3.js

Build Pipeline

Mermaid

Execution State Machine

Mermaid

Sequence Diagram

Mermaid
Personas

The right agents for the right people

Not every developer needs every AI agent. A frontend engineer doesn't need Kubernetes debugging agents. Personas solve this.

When you set up prompticorn, you define which roles exist on your team. Each developer declares their persona and gets a focused, relevant agent set. Universal agents (ask, debug, explain) are always included for everyone.

⚛️

Frontend Engineer

React patterns, accessibility, CSS architecture, browser performance, component design systems, UI testing.

persona: frontend-software-engineer
⚙️

Backend Engineer

API design, database optimization, error handling, service architecture, background workers, data modeling.

persona: backend-software-engineer
🔧

DevOps / Platform

CI/CD pipelines, Kubernetes, infrastructure as code, security hardening, observability, incident response.

persona: devops-engineer
🧪

QA / Testing

Test strategy, coverage analysis, regression patterns, test automation, integration testing, quality gates.

persona: qa-tester
Personas are defined in .prompticorn.yaml

Add roles, remove roles, or create custom ones. Each developer runs prompticorn swap to change their active persona without re-running the full setup.

YAML — .prompticorn.yaml
tool: claude
variant: minimal
language: typescript

personas:
  - frontend-software-engineer
  - backend-software-engineer
  - devops-engineer
  - qa-tester
Supported Tools

Works with the tools your team already uses

Select the tools your team uses during prompticorn init. prompticorn generates the correct config files — automatically, in the right format, at the right paths.

Kilo
Kilo IDE

An AI-native IDE built for agentic development workflows.

Generates.kilo/agents/{agent}.md — YAML frontmatter + Markdown agent files.
Cline
VS Code Extension

An autonomous AI coding agent that runs inside VS Code.

Generates.clinerules — a single drop-in Markdown rules file.
Claude Code
CLI

Anthropic's agentic coding CLI, used directly from the terminal.

Generates.claude/ — full directory of agents, workflows, and skills.
Copilot
GitHub Copilot

GitHub Copilot's agentic coding tool, configured via a single instructions file.

Generates.github/copilot-instructions.md — a single concatenated instructions file.
Cursor
AI Editor

An AI-first code editor built on VS Code with deep agent integration.

Generates.cursor/rules/ + .cursorrules — rules files scoped per agent.
Switching tools

Already set up on Cursor but your team is moving to Claude Code? Run prompticorn switch. Your config is preserved. New files are generated instantly for the new target.

Bash
prompticorn switch
# ? Which AI tool do you want to switch to?
#   > Claude Code
# Regenerating for Claude Code...
# Written: .claude/agents/code.md
# Written: .claude/agents/debug.md
# Done.
Supported AI Tools

Five tools. One config.

Pick your AI coding assistant during prompticorn init — prompticorn writes the right files to the right locations automatically.

Kilo IDE

Kilo IDE

Generates YAML frontmatter plus Markdown agent files for Kilo IDE.

Output: .kilo/agents/{agent_name}.md
YAML frontmatter + Markdown body
Minimal and verbose variants
Persona-filtered
Output Preview
---
name: code
description: Code implementation expert
variant: verbose
---

# Code Agent

You are a senior software engineer...
Cline

Cline

Generates a pure Markdown .clinerules file rendered into Cline's natural-language rule format.

Output: .clinerules
Pure Markdown, prose-based
Minimal and verbose variants
Output Preview
# Code Agent Rules

You are a code implementation expert.

When writing tests, use_skill: testing-strategies
Claude

Claude

Generates the full .claude/ directory with agents, workflows, and skills as Markdown files.

Output: .claude/ directory
Agents, workflows, skills
Minimal and verbose variants
Output Preview
.claude/
  agents/
    code.md
    debug.md
  workflows/
    code.md
  skills/
    testing-strategies.md
GitHub Copilot

Copilot

Generates a single concatenated instructions file for GitHub Copilot's agentic coding tool.

Output: .github/copilot-instructions.md
Single concatenated file
Minimal and verbose variants
Output Preview
---
applyTo: "**"
---

# Code Agent

Apply these instructions when writing code...
Cursor

Cursor

Generates Markdown rules files for Cursor, written to both .cursor/rules/ and .cursorrules.

Output: .cursor/rules/ + .cursorrules
Cursor rules format compliant
Minimal and verbose variants
Output Preview
---
description: Code implementation
globs: ["**/*.py", "**/*.ts"]
---

# Code Agent

You are a senior software engineer...
Supported AI Tools

One config.
Five native outputs.

Pick your AI coding assistant and prompticorn generates the right config files in the right format. No lowest-common-denominator compromises.

Kilo IDE

Kilo IDE

Generates YAML frontmatter + Markdown agent files for Kilo IDE. Files written to .kilo/agents/ automatically.

Output: .kilo/agents/{agent_name}.md
YAML frontmatter generation
Subagent support
Minimal & verbose variants
Output Preview
---
name: code
description: Code implementation expert
subagents:
  - test
  - review
variant: verbose
---

# Code Agent

You are a senior software engineer...
Cline

Cline

Generates a pure Markdown .clinerules file. Rendered into Cline's natural-language rule format.

Output: .clinerules
Pure Markdown output
use_skill invocation pattern
Prose-based rules
Output Preview
# Code Agent Rules

You are a code implementation expert.

When writing tests, use_skill: testing-strategies

## Conventions
Follow the project's established patterns...
Claude

Claude

Generates the full .claude/ directory with agents, workflows, and skills. Populates your entire Claude Code configuration tree.

Output: .claude/ directory
Agents, workflows, skills
Minimal & verbose variants
Persona-filtered output
Output Preview
.claude/
  agents/
    code.md
    debug.md
  workflows/
    code.md
  skills/
    testing-strategies.md
GitHub Copilot

Copilot

Generates a single concatenated instructions file for GitHub Copilot's agentic coding tool.

Output: .github/copilot-instructions.md
Single concatenated file
Version-controlled with repo
Minimal & verbose variants
Output Preview
---
applyTo: "**/*.py"
---

# Code Agent

Apply these instructions when writing
Python code in this repository...
Cursor

Cursor

Generates Markdown rules files for Cursor. Configuration written to both .cursor/rules/ and .cursorrules.

Output: .cursor/rules/ + .cursorrules
Cursor rules format
Per-agent file scoping
Minimal & verbose variants
Output Preview
---
description: Code implementation
globs: ["**/*.py", "**/*.ts"]
---

# Code Agent

You are a senior software engineer.
Follow project conventions strictly...
Get Started

From chaos to aligned in under 5 minutes

Install. Initialize. Commit. Your team is set up.

Deployment

Up and running in minutes

One install command. One init. Production-grade AI agent configuration across five tools, immediately.

Getting Started

Up and running
in minutes.

Bash
pip install prompticorn
Bash
uv add prompticorn

The Init Wizard

Run prompticorn init in your project root. The interactive wizard guides you through every choice:

Bash
$ prompticorn init

? Which AI tools does your team use?
  [x] Claude Code
  [x] Cursor
  [ ] Cline
  [ ] Kilo IDE
  [ ] Copilot

? Primary programming language?  > TypeScript

? Which team roles need to be supported?
  [x] Backend Software Engineer
  [x] Frontend Software Engineer
  [x] DevOps Engineer
  [ ] QA/Tester

Generating configuration...
Written: .claude/agents/code.md
Written: .claude/agents/debug.md
Written: .claude/agents/backend.md
Written: .claude/agents/frontend.md
Written: .cursor/rules/code.mdc
Written: .cursor/rules/debug.mdc
Written: .prompticorn.yaml

Done. Commit .prompticorn.yaml — your team is ready.

Config File

Auto-generated by prompticorn init. Edit directly and run prompticorn update to regenerate.

YAML
# .prompticorn.yaml — auto-generated by `prompticorn init`
# Edit manually and run `prompticorn update` to regenerate

tool: kilo-ide          # kilo-ide | kilo-cli | cline | claude | cursor | copilot
variant: minimal        # minimal | verbose
repo_type: single       # single | monorepo
language: python

personas:
  - backend-software-engineer
  - qa-tester

# Run `prompticorn update` after editing this file
Already set up? Share it with your team.

Commit .prompticorn.yaml. Each developer runs prompticorn init locally to generate their tool-specific configs, tailored to their role. New joiner? Productive in minutes.

CLI Reference

Six commands. Complete lifecycle coverage.

CommandWhat it does
prompticorn initInteractive setup. Configures target tools, active persona, and output paths. Safe to re-run.
prompticorn listLists all registered agents available in the current project configuration.
prompticorn switchSwitches the active AI tool target. Rebuilds all output files for the new tool.
prompticorn swapSwaps the active persona. Re-filters the agent set and regenerates tool-specific configs.
prompticorn updateRegenerates configs from the current .prompticorn.yaml. Non-destructive.
prompticorn validateValidates the config and reports drift, missing files, or integrity issues before they ship.
Teams

Built for teams that take AI seriously

Engineering organizations using prompticorn to align their AI tooling across every developer, every tool, every day.

[Your quote here]
[Name]
CTO, Series A startup
[Your quote here]
[Name]
VP Engineering, B2B SaaS
[Your quote here]
[Name]
Tech Lead, Scale-up
Ready?

Stop configuring.
Start shipping.

One setup. Every tool. Every developer. Aligned.

Get Started Free View on GitHub ↗