feed7.dev
Sign InStart Agent Brain

Trust Methodology

Feed7 separates what a source says from what Feed7 has checked. Trust labels describe provenance and review method; they are not popularity or quality scores.

Official Source

Published by the organization responsible for the product, research, or change.

Source Linked

Feed7 links the underlying source, but has not independently reproduced the claim.

Founder Tested

Feed7 records who tested it, when, in which environment, the steps, result, and limitations.

Repo Verified

The repository, implementation, or release material was reviewed against the claim.

Transcript Verified

A talk or video claim was checked against a source-provided or retrieved transcript.

Needs Review

Useful enough to retain, but a material claim or source relationship still needs editorial review.

Unverified Claim

The source makes a claim that Feed7 cannot currently verify. Treat it as uncertain.

Duplicate Coverage

The item repeats a canonical source without enough new technical material to stand alone.

Uncertainty Rule

If Feed7 has not verified a claim, uncertainty appears near the claim and inside JSON, Markdown, and bundle exports. Agents should preserve it and call out conflicts instead of resolving them by guesswork.