O1-O — deterministic code synthesis operator
Zero-LLM program composition from natural-language intent.
8-color algebraic type system · 7-pass causal inference · 7-layer verification · 9-phase autonomous engagement.
~300ms generation · offline · air-gapped · deterministic
pip install o1op
Quick start
# interactive REPL
o1o
# or run from source
python3 -m o1o_o.o1o_live
What it does
You type intent in plain English. O1-O queries a .causal knowledge graph (~75k triplets, ~1200 code fragments), assembles a working program, runs a 7-layer verification pipeline (compile check → formal verification → evasion scan → OPSEC audit), and hands you deployable code. No model, no API, no network.
O1-O> build a port scanner with banner grabbing for 10.0.0.1
11 pipeline steps, ~300ms, done.
The /engage operator
Full autonomous engagement: recon → tool generation → deployment → post-exploitation → reporting. One command, zero human intervention.
O1-O> /engage 10.0.0.1
What's inside
| Component | Count |
|---|---|
.causal knowledge graph triplets |
~75,000 |
| Code fragments (verified) | ~1,200 |
| MITRE ATT&CK techniques covered | 102 |
| MITRE tactics covered | 14/14 |
| Algebraic type colors | 8 |
| Pipeline verification layers | 7 |
Current state
This works. It generates real code from a real knowledge graph in real time. It's also the first public release and there are rough edges — some fragment selections are imprecise, some intents don't have matching fragments yet, and the engagement operator iterates on failures live (which is a feature, not a bug — you see it retry and adapt).
If something doesn't work, fix it. The fragment registry is plain JSON, the knowledge graph is documented, the pipeline is readable Python. PRs welcome, issues welcome, forks encouraged.
Requirements
- Python ≥ 3.10
- Dependencies:
msgpack,jellyfish,requests,beautifulsoup4
Full documentation
Architecture, papers (9 peer-reviewed at 4 IEEE 2026 conferences + IBM z/OS mainframe validation), math, and the full technical deep-dive:
License
Apache-2.0 · David Tom Foss · dtfoss-dev@proton.me
1 maintainer
Metadata
1 maintainer
Release files for o1op 1.1.0
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 | |
|---|---|---|---|---|
| o1op-1.1.0.tar.gz | 4.9 MB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| o1op-1.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 9.9 MB
Release files / o1op-1.1.0.tar.gz
| Download URL | o1op-1.1.0.tar.gz |
|---|---|
| Size | 4.9 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e6ac5305d3a993980731642b6aafb6da445cdf7a69eb939badb172fa0a1f9951
|
|
BLAKE2b-256 checksum How to use checksums |
9c482b7a7db689c9014cb7232585cef52efe831a4a2a753c67102f86e3cf3bd9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.7
|
Release files / o1op-1.1.0-py3-none-any.whl
| Download URL | o1op-1.1.0-py3-none-any.whl |
|---|---|
| Size | 5.0 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
1777e511408b04aa5d1a992a363adfa0fb0fea264c2b97cfbdf64769e6aade70
|
|
BLAKE2b-256 checksum How to use checksums |
3c91cb7230c363a1935ed5b5640b2d3a8e36b6da9d1f5eadc2e17e0d708b0fa9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.7
|
Security for o1op 1.1.0
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.
