🏢 OpenTower V1.0
5-Layer Multi-Agent Architecture for Zero-Intervention Closed-Loop Rate (ZICR)
From natural language intent to physical execution — an autonomous agent pipeline modeled as the "Empire State Building" of AI orchestration.
Architecture
┌─────────────────────────────────────────────────────────────┐
│ Layer 1: DEFINITION │
│ company.yaml → CEO / Worker / QA roles, prompts, budgets │
├─────────────────────────────────────────────────────────────┤
│ Layer 2: ROUTING │
│ EMPBus (asyncio) — millisecond intent_type dispatch │
├─────────────────────────────────────────────────────────────┤
│ Layer 3: PROTOCOL │
│ EMPPacket JSON — trace_id, action, payload, token_budget │
├─────────────────────────────────────────────────────────────┤
│ Layer 4: AGENTS │
│ CEO (decompose) → Worker (execute) → QA (verify) │
├─────────────────────────────────────────────────────────────┤
│ Layer 5: ACTIONS │
│ shell_execute · github_create_issue · (extensible) │
└─────────────────────────────────────────────────────────────┘
Quick Start
# Install
cd OpenTower
pip install -e ".[dev]"
# Configure LLM (default: LM Studio at localhost:1234)
export OPENAI_BASE_URL=http://localhost:1234/v1
export OPENAI_API_KEY=no-key
# Run
python -m opentower.main
Message Flow
User Input
│
▼
[CEO] ──LLM──> decompose into tasks
│
▼ (task_assign × N)
[Worker] ──Action Registry──> execute
│
▼ (task_result)
[QA] ──static + LLM──> APPROVE / REJECT
│ │
▼ (retry ≤ 2×)
Output
Project Structure
company.yaml # Organization definition (3 agent nodes)
opentower/
├── schema/ # Layer 1 & 3: YAML parser + EMP protocol
├── bus/ # Layer 2: asyncio event bus
├── agents/ # Layer 4: CEO / Worker / QA
├── actions/ # Layer 5: shell, GitHub (extensible)
├── memory/ # SQLite state wall (full trace log)
├── llm/ # OpenAI-compatible client
└── main.py # REPL entry point
Requirements
- Python ≥ 3.11
- An OpenAI-compatible LLM endpoint (LM Studio, Ollama, or cloud)
License
MIT
Metadata
Release files for opentower 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| opentower-1.0.0.tar.gz | 21.7 kB | Without provenance | Details for opentower-1.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| opentower-1.0.0-py3-none-any.whl | Python 3 | none | any | Details for opentower-1.0.0-py3-none-any.whl |
Total release size: 46.1 kB
Release files / opentower-1.0.0.tar.gz
| Download URL | opentower-1.0.0.tar.gz |
|---|---|
| Size | 21.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
dbe247e709af74bed2262a6d8860f9bd6d30c1a6a9006bcb75d97a2bb19a4c74
|
|
BLAKE2b-256 checksum How to use checksums |
9b0b89990360bc20f30fb70e6607bb3134da582346aff2a0bffa7c31fc87fcac
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.12
|
Release files / opentower-1.0.0-py3-none-any.whl
| Download URL | opentower-1.0.0-py3-none-any.whl |
|---|---|
| Size | 24.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0e4bc8693c024f1cf0873d2d6865f1a23d4372a2c8df02377834fa1d2f660a73
|
|
BLAKE2b-256 checksum How to use checksums |
f4b24cbb90f9871e28944b5dafc6fa32cd7da267b10da4ba10ffb30785c46307
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.12
|
Security for opentower 1.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.