A transparent, self-hostable AI news desk you run yourself. Every morning it reads 37 feeds across five continents, clusters the day's stories, decides what matters, writes a bias-labelled briefing, fact-checks its own work, and emails it -- no human edits any issue. Clone it and run your own for a few dollars a day, or read the live instance.
Every choice it makes is inspectable: real subscriber and cost numbers, every source labelled by political bias and factuality, and the code that does it all right here.
- Fully autonomous. No human writes, edits, or approves any issue. The pipeline fetches, curates, writes, and fact-checks itself, then sends.
- Claude never sees a URL. Python assigns opaque article IDs (
A1,A2, ...) and the model curates and writes referencing only those IDs; Python resolves them back to sources afterward. Curation can't be swayed by domain, and a malicious feed can't inject a link into the output. - Five specialized subagents, deterministically orchestrated.
CLUSTER -> RECAP -> SELECT -> WRITE -> COHERENCE, each a file-based Claude Agent SDK subagent that reads and writes JSON, run in a fixed order by Python so the parent context stays small and the run is reproducible. - It fact-checks itself, then repairs itself. A
COHERENCEpass re-reads every headline and summary against its source articles. Anything that fails is first regenerated from its own cited sources, changing as little as possible, and re-checked; only what still fails is dropped, before send rather than after. Every attempt is recorded torepair_log.jsonl. - Cheap clustering by design. Grouping is a deterministic extract-then-join (entities + event + time per article, then a join), not a holistic LLM pass over everything. Lower cost, less drift.
- Evolving story threads. Ongoing stories are tracked across days, so a returning reader sees what changed rather than a fresh fragment.
- Radical transparency. A public stats page shows real subscriber numbers, source balance across the political spectrum, and the AI cost per issue. Every source is labelled by bias and factuality, and the code is right here.
Each story carries a headline, a summary, a why-it-matters note, how the reporting varies across outlets, and the political balance of its sources. See today's issue.
Two components:
- newsroom — the Python pipeline:
fetch -> cluster -> recap -> select -> write (one call per story) -> preheader -> coherence -> repair -> assemble -> render -> email. Stages hand off through JSON files on disk rather than a shared context, so a crashed run resumes and any past run can be replayed from its archived artifacts. Sonnet 5 with adaptive thinking writes and fact-checks; Sonnet 4.6 clusters, selects and repairs; Haiku writes the recap and the inbox preview line. - circulation — a Rust (Axum) web server: the online archive, per-issue pages, the sources and stats pages, story threads, and the "view in browser" links.
Stages and models as of 8344c58. Run figures from run 284 (2026-09-02). That run wrote all stories in one call; the per-story WRITE drawn here had not yet run. Regenerate with make anatomy. Per-stage models and costs: docs/pipeline-anatomy.html.
- Docker
- A Claude subscription (Max or Pro) or API key
- Resend account (free tier: unlimited broadcasts to 1,000 contacts)
git clone https://github.com/SeanLF/claude-rss-news-digest.git
cd claude-rss-news-digest
cp .env.example .envEdit .env with your settings. The required values are:
# Claude authentication (choose one):
CLAUDE_CODE_OAUTH_TOKEN=sk-ant-oat01-... # Subscription: `claude setup-token`
# ANTHROPIC_API_KEY=sk-ant-... # Pay-per-use: console.anthropic.com
# Resend (https://resend.com)
RESEND_API_KEY=re_xxxxxxxx_xxxxxxxxxxxxxxxxxxxx
RESEND_FROM=onboarding@resend.dev
RESEND_AUDIENCE_ID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxxSee .env.example for optional settings (digest name, author, archive URL, etc).
# Test run: fetches articles, generates digest, skips email
docker compose run --rm digest-newsroom --dry-run
# Full run: fetch, curate, email, record
docker compose run --rm digest-newsroomThe database is created automatically on first run.
# Daily at 07:00 UTC
0 7 * * * cd /path/to/news-digest && docker compose run --rm digest-newsroom >> data/cron.log 2>&1docker compose up -d digest-circulation
# Browse at http://localhost:808037 feeds from 30 outlets across five continents, spanning the political spectrum. Bias, factual reporting and credibility are taken from Media Bias/Fact Check, read per outlet so each rating traces to one published assessment; 26 of 37 feeds rate High or Very High for factual reporting, 10 Mostly Factual or Mixed, and one (Hacker News) is unrated. Every source is shown with its bias and factuality on the live sources page. See newsroom/sources.json for the full list.
Roughly a few dollars a day in API-equivalent cost (Sonnet for the reasoning stages, Haiku for recap). The live stats page shows the current per-issue number.
| Problem | Fix |
|---|---|
| No digest generated | Check data/digest.log; verify auth with docker compose run --rm digest-newsroom claude --version |
| Email not sending | Verify RESEND_API_KEY and RESEND_FROM in .env; test with docker compose run --rm digest-newsroom --test-email you@example.com |
| Container issues | docker compose build --no-cache |
- Production deployment -- docs/DEPLOYMENT.md
- Architecture and dev context -- CLAUDE.md
- All CLI flags --
docker compose run --rm digest-newsroom --help
PolyForm Noncommercial License 1.0.0 -- free to use, modify, and share for any noncommercial purpose (personal, research, education, nonprofits). Commercial use, including by for-profit organizations, is not permitted. This is a source-available licence, not an OSI open-source one.