Verification register AI & Agents
Current reading
The readiness gap, in one scan
AI-assisted assembly · derived results
- Claimed
- 80
- Reported
- 72
- Verified
- 61
- Gap
- +19
Public ambition and stated capability
Observed practitioner reporting
Independently supported evidence
Claimed minus verified
Evidence strength Critical
Decision
What the current evidence supports
Human editorial judgment · 2026-06-28
Adopt with guardrails
- Why
- Officially production-blessed, now embedded in the 2026-07-28 spec, with broad client support and real shipped apps, but documented cross-host CSP, streaming-readiness, and SDK-coverage defects mean it needs guardrails and multi-host testing, not blind adoption.
- Next
- Prototype one ephemeral, single-purpose ui:// widget behind a host-abstraction layer; test it explicitly on Claude.ai AND ChatGPT and verify CSP frameDomains before relying on external-origin iframes; prefer TypeScript for the UI bundle given the SDK gap; avoid embedding full apps.
Constraints
Blockers
No named blocker is present in the current public projection.
Evidence summary
Derived counts
AI-assisted assembly
- Total
- 7
- Tier 1
- 0
- Tier 2
- 2
- Tier 3
- 5
- Supports
- 4
- Contradicts
- 3
- Context
- 0
- Latest observed
- 2026-07-28
Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.
Publication record
Revisions
Initial public reading
This is the initial public reading. No earlier readiness change is recorded.