Specification compiler for AI agent orchestration
Project description
Blueprint
Specification compiler for AI agent orchestration.
Blueprint transforms natural language goals into structured, compilable, interface-first contracts that any AI agent can execute with deterministic results.
Why Blueprint?
As AI coding agents become mainstream, the bottleneck shifts from "knowing programming languages" to "articulating clear, executable instructions." Natural language is expressive but ambiguous — and ambiguity compounds catastrophically with parallel agents.
Blueprint solves this by compiling goals into structured specifications with:
- Interface contracts between tasks (input/output types)
- Dependency graphs (DAG structure for parallelization)
- Built-in verification (test commands, acceptance criteria)
- Human-in-the-loop signals (explicit pause points)
Installation
pip install blueprint-ai
Or with LLM support for generation features:
pip install blueprint-ai[llm]
Requirements
- Python 3.11+
ANTHROPIC_API_KEYenvironment variable (for generation features)
Quick Start
Generate a Blueprint from a Goal
export ANTHROPIC_API_KEY="your-key"
blueprint generate "Build a user authentication system with JWT tokens"
from blueprint.generator import generate_blueprint
roadmap = generate_blueprint(
goal="Build a user authentication system with JWT tokens",
context="Using FastAPI and PostgreSQL"
)
print(roadmap)
Validate an Existing Blueprint
blueprint validate my_roadmap.md
Execute a Blueprint (Dry Run)
blueprint execute my_roadmap.md --dry-run
⚠️ Important: Blueprint is a Compiler, Not a Template
DO NOT manually write Blueprint format documents.
Blueprint's value comes from LLM-powered decomposition, interface inference, and validation. Manual generation bypasses these guarantees.
Always use the blueprint generate command or Python API.
Documentation
- Blueprint Specification — Format reference
- System Architecture — Internal design
- API Interface — Integration guide
License
Apache 2.0 — See LICENSE for details.
Contributing
Contributions welcome! Please read CONTRIBUTING.md before submitting PRs.
Blueprint — "Goals become roadmaps"
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file blueprint_ai-1.2.0.tar.gz.
File metadata
- Download URL: blueprint_ai-1.2.0.tar.gz
- Upload date:
- Size: 71.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4ad19b2667309fc853d1d8d70b4a7f8e675c4faf1d293a5697a5d766e7d9f7bb
|
|
| MD5 |
89b65e47cad5b6c5eb6a1cb0788d5372
|
|
| BLAKE2b-256 |
43430c1026155d741ac4de43d05a36fd6f357f6ae84dbca10a7d4cbd46a37bcd
|
File details
Details for the file blueprint_ai-1.2.0-py3-none-any.whl.
File metadata
- Download URL: blueprint_ai-1.2.0-py3-none-any.whl
- Upload date:
- Size: 51.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f574ec571d0cfb404c6c9f1c1c3b31ae47dfd365c3987fae11fc18856c644ae2
|
|
| MD5 |
9ab6bca406e40136918d8af204859e63
|
|
| BLAKE2b-256 |
d7c38fd82ef8bf710c9c3d1e07985226d1a164615915f7534f256a911f644d99
|