# feed7.dev

> Source-backed AI engineering signals, a maintained Atlas, and agent-ready context for software builders.

Feed7 separates source facts from editorial judgment. Every public signal links to its original source and exposes trust status, uncertainty, practical implication, and agent-ready context.

## Primary machine-readable surfaces

- [JSON Feed](https://feed7.dev/feed.json): current enriched briefs and canonical posts
- [RSS Feed](https://feed7.dev/feed.xml): current enriched briefs and canonical posts
- [Latest weekly issue](https://feed7.dev/weekly/2026-07-07): finite editorial selection
- [Latest weekly JSON](https://feed7.dev/weekly/latest.json): structured issue and signals
- [Latest weekly Markdown](https://feed7.dev/weekly/latest.md): agent-ready issue context
- [OpenAPI document](https://feed7.dev/openapi.json): public read-only HTTP contract
- [Atlas](https://feed7.dev/atlas): maintained topic pages and evidence — every taxonomy node is a page
- [Atlas graph](https://feed7.dev/atlas/graph.json): all nodes plus approved, source-referenced edges
- [Trust methodology](https://feed7.dev/trust): verification labels and uncertainty rules

## URL patterns

- HTML signal: https://feed7.dev/p/{slug}
- JSON signal: https://feed7.dev/p/{slug}.json
- Markdown signal: https://feed7.dev/p/{slug}.md
- Atlas JSON: https://feed7.dev/atlas/{topic}.json — evidence includes the harvested corpus (corpus_evidence)

## Use policy

Prefer the linked original source for source claims. Use Feed7 for its editorial synthesis, trust labeling, cross-source context, practical implications, and agent-ready bundles. Do not infer verification beyond the explicit trust status and uncertainty fields.