Self-hosted CVE intelligence for analysts who need answers, not another raw feed

Live demo · Documentation · Quick start · Screenshots

What is BRIEFR?

BRIEFR is a self-hosted CVE intelligence dashboard. It pulls public vulnerability and threat feeds into your PostgreSQL database, then gives you one dark-mode UI to work through them: morning brief, searchable feed, IOC lookup, investigation graph, advisories, and detection tooling.

You are not clicking through NVD on every page load. Schedulers sync upstream sources on a schedule; the UI reads what is already stored locally.

Why BRIEFR?

• Save time on triage — Operational Priority (P1–P4), threat score, and stack relevance surface what matters first instead of a flat severity sort.
• One place to work — Brief, feed, drawer detail, IOC enrichment, stored-intel graph, and Forge hunt packs share the same session and watchlist.
• Explainable scoring — Rule-based prioritization and correlation; optional LLM narration at the edges, not a black-box risk number.
• Detection output — Sigma, SIEM snippets, and YARA from a local SigmaHQ mirror plus BRIEFR class templates tied to CWE/ATT&CK context.
• Your data stays yours — Self-hosted Postgres, Apache 2.0, no vendor SaaS lock-in.

Who is it for?

• SOC and vulnerability analysts who need a daily queue and deep CVE context
• Detection engineers mapping CVEs to ATT&CK and exporting hunt content
• Small security teams who want one tool instead of five browser tabs and spreadsheets
• Operators who need Postgres backups, scheduler health, and API key management on their own infra

Not a scanner or ASM product — BRIEFR prioritizes known CVEs against your stack; it does not discover assets on your network.

What you get

Try without installing: briefrdemo.projectjupiter.in — same analyst shell over fixture data.

How data reaches you

1. Ingest — Background jobs sync NVD, CISA KEV, EPSS, MITRE, OTX, RSS, and other sources into PostgreSQL (respecting each provider’s rate limits).
2. Store — CVE rows, enrichment mirrors, correlation artifacts, and optional pgvector embeddings live in your database.
3. Work — The React UI reads from Postgres. IOC lookups and optional LLM tasks call outbound APIs only when you ask.
4. Operate — Admin surfaces show feed freshness, scheduler jobs, backups, and API key health.

For architecture depth, see How it works and System design.

Quick start

Local try-out (SQLite fallback — good for a first look):

git clone https://github.com/Soldier0x0/briefr.git && cd briefr
cd backend && python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements-dev.txt && cp .env.example .env
DATABASE_URL="" BRIEFR_REQUIRE_POSTGRES=0 uvicorn main:app --host 0.0.0.0 --port 8000

cd ../frontend && npm install && npm run dev   # http://localhost:5173

Open the UI and complete first-run admin setup.

Production path — PostgreSQL 16 + pgvector (pgvector/pgvector:pg16), then the install script:

# After Postgres is running — see docs/SELF_HOST.md §3
bash deploy/briefr-install.sh
curl -s http://127.0.0.1:8000/api/health | python3 -m json.tool

Online docs: docs.projectjupiter.in

Screenshots

From a self-hosted deployment (dark theme, BRIEFR accent #e85533).

Documentation

License

Apache License 2.0 — see LICENSE. Security reports: SECURITY.md.

Copyright © 2026 Sai Harsha Vardhan.