https://github.com/mizcausevic-dev/evidence-ranking-engine
Python FastAPI engine for ranking evidence packets by trust, freshness, citations, and contradiction pressure.
https://github.com/mizcausevic-dev/evidence-ranking-engine
analytics backend citations decision-intelligence evidence-ranking fastapi governance portfolio python trust-scoring
Last synced: 2 days ago
JSON representation
Python FastAPI engine for ranking evidence packets by trust, freshness, citations, and contradiction pressure.
- Host: GitHub
- URL: https://github.com/mizcausevic-dev/evidence-ranking-engine
- Owner: mizcausevic-dev
- Created: 2026-05-15T17:17:45.000Z (19 days ago)
- Default Branch: main
- Last Pushed: 2026-05-17T05:12:55.000Z (17 days ago)
- Last Synced: 2026-05-17T07:27:07.364Z (17 days ago)
- Topics: analytics, backend, citations, decision-intelligence, evidence-ranking, fastapi, governance, portfolio, python, trust-scoring
- Language: HTML
- Homepage: https://kineticgain.com/
- Size: 1.38 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
Awesome Lists containing this project
README
# Evidence Ranking Engine
Python and FastAPI engine for ranking evidence packets by trust, freshness, citation density, and contradiction pressure before recommendations become executive-facing advice.
## Why it matters
Decision systems usually fail by over-trusting stale or weakly cited evidence. This repo turns evidence quality into an operator-visible surface: which packets are safe to lead with, which ones are aging, and which ones are under contradiction pressure before a recommendation chain hardens around them.
## Screenshots
## What it does
- ranks evidence packets across trust score, citation depth, freshness, and contradiction count
- exposes conflict lanes that need review before a decision memo inherits them
- shows owner stewardship lanes for evidence quality upkeep
- provides a prompt-fit API to re-rank evidence against a fresh operating question
## Local run
```powershell
cd evidence-ranking-engine
py -3.11 -m venv .venv
.\.venv\Scripts\python.exe -m pip install -r requirements.txt
.\.venv\Scripts\python.exe -m app.main
```
If `5026` is already taken:
```powershell
$env:PORT = "5031"
.\.venv\Scripts\python.exe -m app.main
```
Then open:
- `http://127.0.0.1:5026/`
- `http://127.0.0.1:5026/evidence-board`
- `http://127.0.0.1:5026/conflicts`
- `http://127.0.0.1:5026/owners`
- `http://127.0.0.1:5026/docs`
## Validation
```powershell
.\.venv\Scripts\python.exe -m unittest discover -s tests
.\.venv\Scripts\python.exe scripts\run_demo.py
.\.venv\Scripts\python.exe scripts\smoke_check.py
.\.venv\Scripts\python.exe scripts\render_readme_assets.py
```
## API routes
- `GET /api/dashboard/summary`
- `GET /api/evidence`
- `GET /api/evidence/{evidence_id}`
- `GET /api/conflicts`
- `GET /api/owners`
- `GET /api/sample`
- `POST /api/analyze/rank`
## Repo layout
- `app/main.py` FastAPI routes and API surface
- `app/services/evidence_ranking_service.py` ranking, conflict scoring, and owner lane logic
- `app/render.py` HTML control surfaces and static proof generation
- `app/data/sample_evidence_data.py` seeded evidence packets and dashboard data
- `docs/architecture.md` system structure and ranking model
- `docs/ORIGIN.md` why the product exists