🏛️ Agba Engine (`agba-engine`)

Version: 5.2.4
Author & Lead Architect: Aruna Olanrewaju Kabiru
License: RAIL-Cultural-Heritage-v1.0

agba-engine is a cross-modal vision and diacritic-preserving text retrieval framework engineered to index Global Cultural Heritage, Indigenous Scripts, and Tonal Orthographies without semantic stripping, Unicode degradation, or sub-token drift.

🌍 The Mission: Preserving Global Heritage

Standard LLMs and vector search engines strip diacritics, flatten tonal markers, and misclassify rare indigenous vocabulary. Agba Engine provides an open, diacritic-proof dual-retrieval architecture that preserves native orthographies in their full canonical purity while integrating neural zero-shot vision search.

📦 Quickstart & Global Developer Workflow

Installation

pip install agba-engine==5.2.4

Initializing Engine & Executing Searches

from agba_engine import AgbaProductionEngine

# Initialize dual engine (loads bundled ground-truth corpus & static SigLIP embeddings)
engine = AgbaProductionEngine()

# 1. Deterministic Text Retrieval (Diacritic & ASCII-Folded Parity)
text_results = engine.search_by_text("Opon Ifa", top_k=1)
print(f"Retrieved Entity: {text_results[0]['id']} | Score: {text_results[0]['score']}")

# 2. Cross-Modal Neural Vision Search (SigLIP + OOD Noise Guardrail)
vision_results = engine.search_by_vision("https://example.com/yewa_beads.jpg", top_k=3, ood_threshold=0.0600)
if vision_results["status"] == "MATCH_FOUND":
    for hit in vision_results["results"]:
        print(f"Rank {hit['rank']}: {hit['id']} ({hit['class']}) — Cosine: {hit['raw_cosine']}")

🛠️ API Reference

Method	Parameters	Return Type	Description
`search_by_text()`	`text_query: str`, `top_k: int = 5`	`List[Dict]`	Executes deterministic token retrieval with ASCII-folding (`0.9600`) & exact diacritic parity (`1.0000`).
`search_by_vision()`	`image_input: Union[str, Image.Image]`, `top_k: int = 5`, `ood_threshold: float = 0.0600`	`Dict`	Cross-modal SigLIP visual search against pre-computed static embeddings with OOD noise rejection.

🏛️ Enterprise & Institutional API Access (Museums & Archives)

agba-engine provides specialized integration hooks for cultural institutions, digital heritage archives, and global museum databases seeking high-fidelity index retrieval for native orthographies, 3D artifacts, and historical vocalization assets.

Audio & Vocalization Pipeline: Endpoints for indexing pitch-contour models, tonal pronunciation vectors, and spoken indigenous archives.
Museum Corpus Sync: Direct ingestion pipelines for institutional digital asset management (DAM) platforms.

For API access keys, custom corpus integration support, or institutional partnerships, reach out directly or open a request on our Hugging Face Space:

GitHub Repository: Agba-Engine on GitHub
Hugging Face Space: Agba Engine Live Demo & Community
Direct Contact: only1mooseylion@gmail.com

Metadata

Release files for agba-engine 5.2.4

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for agba-engine 5.2.4
File	Size	Uploaded	Provenance
agba_engine-5.2.4.tar.gz	2.8 MB	Sep 12, 2026	Without provenance	Details

Built distribution (wheel)

Table of built distributions (wheels) for agba-engine 5.2.4
File	Interpreter	ABI	Platform	Reset
agba_engine-5.2.4-py3-none-any.whl 2.8 MB Sep 12, 2026 Without provenance	Python 3	none	any	Details

Total release size: 5.6 MB

Release files / agba_engine-5.2.4.tar.gz

Download URL	agba_engine-5.2.4.tar.gz
Size	2.8 MB
Tags	Source
SHA-256 checksum How to use checksums	`f2906aa49b53e34ee9985986bb0d9a2455d01a3f6c3432f3ed77b0b362ed6bc6`
BLAKE2b-256 checksum How to use checksums	`58800ddc38f02c4836db8581df677b8f69c3d5a76af7681a058f89e9111e45af`
Upload date	Sep 12, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.15`

Release files / agba_engine-5.2.4-py3-none-any.whl

Download URL	agba_engine-5.2.4-py3-none-any.whl
Size	2.8 MB
Tags	Python 3
SHA-256 checksum How to use checksums	`f2e09ab6c8bc5f605d7d7b61e08a8cbcf04066c7eaa76eb5b8258b4ef61d2f16`
BLAKE2b-256 checksum How to use checksums	`e37219bc10b251b865dcfd5ec64b3b02803bb3ae63476d8555e634c069e2f010`
Upload date	Sep 12, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.15`

Security for agba-engine 5.2.4

Report project as malware

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.