Ingest, enrich and search legislative records through an agentic AI pipeline — with classification, entity extraction, grounded Q&A, verified citations and role-based access for members, ministers and citizens.
From upload to insight — every document is enriched, indexed and made searchable with sources you can trust.
Every document flows through 10 live agent stages — OCR, classification, NER, embedding, summarisation — streamed in real time.
Keyword, semantic, multilingual and natural-language Q&A fused and reranked into one precise result set.
Every AI answer is checked for faithfulness — each claim is confirmed against its source and highlighted on the exact page.
A hard security boundary — members, ministers, speakers and citizens each see only what their role permits.
English plus regional languages, first-class — cross-lingual embeddings retrieve the right document regardless of query language.
Search behaviour, latency, corpus health and top entities — the whole platform at a glance.
Each microservice runs as an independent agent, with live progress for every stage.
Scan · dedup · route
OCR / text extract
Classify + metadata
Entities + chunking
Embed + vector index
Summary + key points
Store + publish
Co-Pilot, speeches, constituency intelligence
Bill drafting, replies, assurance tracking
Rules engine, admissibility, precedents
Live floor, timing, translation
Published records, bill summaries, search
Sign in with your platform account, or explore public legislative records as a citizen.
Browse, filter and manage the legislative corpus
Keyword · Semantic · Multilingual · NL Q&A — fused & reranked.
Add many PDFs or images at once. Each file runs through the ACTIGEN pipeline sequentially.
No files in queue — add PDFs or images above.
Upload a document to watch each agent stage complete in real time.
Loading analytics…
Access the legislative document platform
New citizen? Create an account