Skip to main content

WronAI Hypervisor — orchestrator and control plane for AI desktop agents, NLP-to-URI resolution, koru drivers, and virtualized execution

Project description

Resource Agent Meta-Factory v0.1

AI Cost Tracking

PyPI Version Python License AI Cost Human Time Model

  • 🤖 LLM usage: $1.7735 (5 commits)
  • 👤 Human dev: ~$288 (2.9h @ $100/h, 30min dedup)

Generated on 2026-06-14 using openrouter/qwen/qwen3-coder-next


Generator, validator and repairer for contract-first thin agents.

This package extends the earlier resource-agent-factory with a meta-agent that can create agent specifications from prompts, validate them, repair common mistakes and generate working FastAPI agents.

Core idea

Prompt / request
      ↓
Meta-agent
      ↓
contracts/agents/*.yaml
      ↓
Validator + safe repair
      ↓
Agent Factory
      ↓
agents/generated/<agent>/
      ↓
Tests + contract hash verification

The LLM or user creates the contract proposal. The deterministic generator creates the code.

Quick start

Install dependencies:

pip install -e '.[dev]'

Validate existing contracts:

make validate

Generate agents:

make generate

Verify generated agents:

make verify

Run tests:

make test

Meta-agent workflow

Create a new agent from a prompt:

python -m meta_agent.cli pipeline "Stwórz agenta do obsługi zamówień z odczytem zamówienia, historią i tworzeniem zamówienia"

Run the meta-agent HTTP API:

make run-meta-agent

Then call:

curl -X POST http://localhost:8200/pipeline/from-prompt \
  -H 'Content-Type: application/json' \
  -d '{"prompt":"Stwórz agenta do obsługi faktur z odczytem faktury, historią i tworzeniem faktury"}'

Important rule

Do not edit agents/generated/ manually.

Change:

contracts/agents/*.yaml

Then regenerate.

Resource Agent Hypervisor v0.5

Ta wersja dodaje osobne paczki uri3 i nl2uri.

  • uri3 = parser, normalizer, resolver, scanner, graf zależności i walidator URI Tree.
  • nl2uri = natural language/query/prompt -> URI Tree, korzystając z modelu URI i walidacji z uri3.

Hypervisor nie skanuje usług samodzielnie i nie generuje domeny bezpośrednio z promptu. Robi to przez uri3 i nl2uri.

prompt -> nl2uri -> URI Tree -> uri3 validation/graph -> Domain Pack -> Agent Factory -> generated thin agent

Instalacja

pip install -e .[dev]

Przykład

nl2uri generate --no-llm -p "generuj mape pogody dwa tygodnie do przodu w oparciu o miejscowosc i odpowiedni model przewidujacy pogode, generuj widok w formie html pod adresem url" --out domains/weather_map/uri_tree.yaml
uri3 validate-tree domains/weather_map/uri_tree.yaml
uri3 graph domains/weather_map/uri_tree.yaml
nl2a generate --no-llm -p "generuj mape pogody dwa tygodnie do przodu w oparciu o miejscowosc i odpowiedni model przewidujacy pogode, generuj widok w formie html pod adresem url"

Zasada odpowiedzialności

uri3 = adresowanie, discovery, routing, graf URI
nl2uri = język naturalny -> URI Tree
hypervisor = registry, policy, deployment, lifecycle
agent factory = generowanie kodu cienkiego agenta
domain pack = logika domenowa

Documentation

  • docs/META_AGENT.md — meta-agent usage.
  • docs/ARCHITECTURE_META_FACTORY.md — architecture.
  • docs/AUTO_EVOLUTION_PIPELINE.md — controlled auto-evolution.
  • docs/CONTRACTS.md — YAML contract format.
  • docs/GENERATOR.md — generator details.
  • docs/DEPLOYMENT.md — deployment notes.

License

Licensed under Apache-2.0.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hypervisor-0.1.2.tar.gz (52.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

hypervisor-0.1.2-py3-none-any.whl (70.1 kB view details)

Uploaded Python 3

File details

Details for the file hypervisor-0.1.2.tar.gz.

File metadata

  • Download URL: hypervisor-0.1.2.tar.gz
  • Upload date:
  • Size: 52.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.7

File hashes

Hashes for hypervisor-0.1.2.tar.gz
Algorithm Hash digest
SHA256 33deb5fd337c2029301117c9e95775468dfdb9753bec35ad22b873d0601b5f2c
MD5 fc174abc68a2c9f35c54433016aaed99
BLAKE2b-256 e75718c112cfa3b8015a0a562720055a2307b78644b7bf44b3b330b5f3db88ea

See more details on using hashes here.

File details

Details for the file hypervisor-0.1.2-py3-none-any.whl.

File metadata

  • Download URL: hypervisor-0.1.2-py3-none-any.whl
  • Upload date:
  • Size: 70.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.7

File hashes

Hashes for hypervisor-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 858303f3707c375298a78b716c200c4a28b72e5b62313a137b4f1883801be911
MD5 777632c44864f98aa4d3058ebf364691
BLAKE2b-256 631c6736dbde0d2647e1618dd2fec917e60a8cb30cb3ca14dda7e6dbf3a4c140

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page