Two-step agent-insurability evidence flow: score an agent execution evidence bundle for underwriter-facing evidence completeness, then build a two-sided replay-challenge dossier for a disputed execution_claim, optionally testing a warranty KPI breach.
score_agent_insurability_evidence{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "score_agent_insurability_evidence",
"arguments": {
"underwriter_profile": "aiuc",
"execution_claims": [
{
"execution_hash": "sha256:",
"human_oversight": true
}
],
"receipts": [
{
"receipt_hash": "sha256:"
}
]
}
},
"id": 1
}
build_claim_dispute_bundle{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "build_claim_dispute_bundle",
"arguments": {
"execution_claim": {
"execution_hash": "sha256:",
"tool_id": "art-251-compute-parametric-trigger-payout",
"receipts": [
{
"receipt_hash": "sha256:"
}
]
},
"challenge": {
"digest": "sha256:"
}
}
},
"id": 2
}
Copy this paragraph into Claude, OpenClaw, or any MCP-aware agent to run this whole workflow on the server and check one field that answers a real question.
Question: A policyholder disputes our denial and points at the uptime warranty. Does the assembled bundle prove a KPI breach?
Run the AINumbers MCP tool `run_chain` with {"chain":"claim-dispute-bundle-assembly"} and read `art-307-claim-dispute-bundle-builder.kpi_breach.breached`.
Re-run with `inputs` for `art-307-claim-dispute-bundle-builder`, reusing the values that result echoed, with `warranty_kpi_breach` changed so measured_metric is 99.9, above the 99.5 threshold. Compare the same field.
Verify: call `verify_execution_hash` (https://mcp.ainumbers.co/mcp) with `claimed_hash` set to `composite_execution_hash` and the full `composite_artifact`.
Ledger, for a human re-check: https://ledger.ainumbers.co/
PII rule: send synthetic or anonymised inputs only. The MCP server runs these kernels and logs no payloads.
Chain page: https://ainumbers.co/chaingraph/chains/claim-dispute-bundle-assembly.html