Verification register AI & Agents
Current reading
The readiness gap, in one scan
AI-assisted assembly · derived results
- Claimed
- 70
- Reported
- 66
- Verified
- 59
- Gap
- +11
Public ambition and stated capability
Observed practitioner reporting
Independently supported evidence
Claimed minus verified
Evidence strength Growing
Decision
What the current evidence supports
Human editorial judgment · 2026-06-28
Adopt with guardrails
- Why
- It is the most directly relevant public benchmark for our MCP-heavy architecture, but external-key fragility and an unsaturated, judge-coupled metric mean it should inform internal gates, not be a hard public claim.
- Next
- Pilot MCP-Bench's multi-server task set against our own MCP tool fleet (apps/mcp-*) as an offline regression gate; pin the judge model and cache external API fixtures to make reruns deterministic. Reuse its Tool Name Validity Rate / Schema Compliance Rate as cheap rule-based checks.
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
- 3
- Tier 3
- 4
- Supports
- 4
- Contradicts
- 2
- Context
- 1
- Latest observed
- 2025-09-01
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.