Async on-prem LLM-powered structured information extraction microservice

Find a file

goldstein b2ff27c1ca Some checks are pending tests / test (push) Waiting to run Details feat(segmentation): SegmentIndex + prompt-text formatter (spec §9.1) (#6 ) SegmentIndex lands.		2026-04-18 08:54:02 +00:00
.forgejo/workflows	ci: run on every push (not just main) so feat branches also get CI	2026-04-18 10:40:44 +02:00
docs	Implementation plan for ix MVP	2026-04-18 10:34:30 +02:00
src/ix	feat(segmentation): SegmentIndex + prompt-text formatter (spec §9.1)	2026-04-18 10:53:46 +02:00
tests	feat(segmentation): SegmentIndex + prompt-text formatter (spec §9.1)	2026-04-18 10:53:46 +02:00
.env.example	feat(scaffold): project skeleton with uv + pytest + forgejo CI	2026-04-18 10:36:43 +02:00
.gitignore	feat(scaffold): project skeleton with uv + pytest + forgejo CI	2026-04-18 10:36:43 +02:00
.python-version	feat(scaffold): project skeleton with uv + pytest + forgejo CI	2026-04-18 10:36:43 +02:00
AGENTS.md	Initial design: on-prem LLM extraction microservice MVP	2026-04-18 10:23:17 +02:00
pyproject.toml	feat(scaffold): project skeleton with uv + pytest + forgejo CI	2026-04-18 10:36:43 +02:00
README.md	Initial design: on-prem LLM extraction microservice MVP	2026-04-18 10:23:17 +02:00
uv.lock	feat(scaffold): project skeleton with uv + pytest + forgejo CI	2026-04-18 10:36:43 +02:00

README.md

InfoXtractor (ix)

Async, on-prem, LLM-powered structured information extraction microservice.

Given a document (PDF, image, text) and a named use case, ix returns a structured JSON result whose shape matches the use-case schema — together with per-field provenance (OCR segment IDs, bounding boxes, cross-OCR agreement flags) that let the caller decide how much to trust each extracted value.

Status: design phase. Implementation about to start.

Full reference spec: docs/spec-core-pipeline.md (aspirational; MVP is a strict subset)
MVP design: docs/superpowers/specs/2026-04-18-ix-mvp-design.md
Agent / development notes: AGENTS.md

Principles

On-prem always. LLM = Ollama, OCR = local engines (Surya first). No OpenAI / Anthropic / Azure / AWS / cloud.
Grounded extraction, not DB truth. ix returns best-effort fields + provenance; the caller decides what to trust.
Transport-agnostic pipeline core. REST + Postgres-queue adapters in parallel on one job store.