A nine-stage agent trade evidence workflow over caller-declared inputs. Step 1 validates the trading agent's on-behalf-of mandate, its subject, scope, intent and validity window, and step 2 then applies a fixed ten-rule authorization order to the declared mandate, budget, rail state and proposed request. Step 3 aggregates the declared oracle submissions into the single price an oracle network would publish, and step 4 then publishes the head event that sequence-numbers the mark-price print stream. Step 5 slices the declared order into VWAP, TWAP and participation-of-volume schedules and decomposes implementation shortfall; step 6 then scores every declared fill for price improvement against the NBBO at execution, and step 7 weighs the declared per-venue figures into a monitoring evidence pack with an overall verdict. Step 8 validates the agent's audit-trail records for completeness and chain-link conformance, and step 9 then aggregates the session into one Merkle-root receipt with inclusion proofs. Each step computes from its own declared inputs; the order is analytical, not a statutory sequence, and the chain certifies nothing.
validate_agent_obo_mandate{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "validate_agent_obo_mandate",
"arguments": {}
},
"id": 1
}
authorize_payment_against_mandate{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "authorize_payment_against_mandate",
"arguments": {}
},
"id": 1
}
oracle_price_aggregation{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "oracle_price_aggregation",
"arguments": {}
},
"id": 1
}
publish_market_mark_head{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "publish_market_mark_head",
"arguments": {}
},
"id": 1
}
compute_algo_execution_schedule_simulator{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "compute_algo_execution_schedule_simulator",
"arguments": {}
},
"id": 1
}
recompute_best_execution{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "recompute_best_execution",
"arguments": {}
},
"id": 1
}
compute_best_execution_evidence_pack{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "compute_best_execution_evidence_pack",
"arguments": {}
},
"id": 1
}
validate_agent_audit_trail{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "validate_agent_audit_trail",
"arguments": {}
},
"id": 1
}
aggregate_execution_receipts{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "aggregate_execution_receipts",
"arguments": {}
},
"id": 1
}
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: Our agent traded on a user's behalf. Does its mandate validate on subject, scope and expiry, or does the validator refuse it?
Run the AINumbers MCP tool `run_chain` with {"chain":"agent-trade-evidence"} and read `art-151-agent-obo-mandate-validator.verdict`.
Re-run with `inputs` for `art-151-agent-obo-mandate-validator`, reusing the values that result echoed, with `now_unix` set to 10000000000. 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/agent-trade-evidence.html