AI governance and assurance, not lifecycle-bound. Identifier handoff (subject_hash), not a computed value passed on. Two-step quarterly evidence flow: bind a sealed artifact into a subject identifier, then compose the quarter's testing-evidence pack tied to that subject. Stop condition before step 2: proceed only when step 1 reports binding_complete true and producer_pinned true; a pack built over an incomplete or unpinned subject is labelled not attested.
bind_attested_subject{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "bind_attested_subject",
"arguments": {}
},
"id": 1
}
build_agent_test_evidence{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "build_agent_test_evidence",
"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: Is this quarter's test evidence tied to one identified, pinned subject, or is the pack not attested?
Run the AINumbers MCP tool `run_chain` with {"chain":"attested-subject-to-quarterly-test-evidence-pack"} and read `art-378-quarterly-test-evidence-composer.ha_evidence_bundle`.
Re-run with `inputs` for `art-378-quarterly-test-evidence-composer`, reusing the values that result echoed, with `subject_hash` changed so remove subject_hash and watch ha_evidence_bundle go null. 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/attested-subject-to-quarterly-test-evidence-pack.html