Evidence · first-party tested

The lead agent can label the lead but does not expose a numeric confidence score; the tested run showed only "Medium-Fit" with no 6/10-style value or criteria breakdown.

⚠ StruggledRelevance AI →
Output — unretouched
tool output
The observation

The lead agent can label the lead but does not expose a numeric confidence score; the tested run showed only "Medium-Fit" with no 6/10-style value or criteria breakdown.

Criterion: Structured output · Scenario: Borderline B2B lead qualification for QuickCart India (group: lead-qualification)
Query this
get_evidence({
  tool: "relevance-ai",
  scenario: "lead-qualification"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, other tools
Dust ✓ WorkedPickaxe ⚠ StruggledZapier AI Agents ✓ Worked
Real outputs, no retouching · every cell queryable via API & MCP · aidemos.com