Calibration runs exercise each verifier in the Source Network against a seed corpus of claims whose ground truth has been established independently. Precision is the share of positive verdicts that were correct; recall is the share of real positives that were caught. F1 combines the two. A verifier with a missing API key returns unverifiable on every claim and therefore scores n/a in the table below; the row is committed to make the key-missing code path legible, not because it represents the verifier's calibrated reliability. See the calibration hub for the run index and the methodology protocol. The underlying per-claim verdicts are published as raw_verdicts.json so the aggregation here is reproducible.

Source Network Calibration Report

• Corpus version: 0.1
• Corpus seeded: 2026-04-16
• Run started: 2026-11-05T14:25:53.296026+00:00
• Total claims: 33
• Stale ground-truth claims: 33 (see list at end)

Per-provider results (all claims)

Claim-level detail

Stale ground-truth claims

These claims have an as_of_date older than 90 days. Re-verify the primary source before treating their results as calibration signal.

• sec-001
• sec-002
• sec-003
• sec-004
• sec-005
• fred-001
• fred-002
• fred-003
• fred-004
• fred-005
• rc-001
• rc-002
• rc-003
• rc-004
• rc-005
• wb-001
• wb-002
• wb-003
• wb-004
• av-001
• av-002
• av-003
• av-004
• wa-001
• wa-002
• wa-003
• wa-004
• wiki-001
• wiki-002
• wiki-003
• wiki-004
• wiki-005
• wiki-006