test: retain accepted Leo challenger proof

This commit is contained in:
twentyOne2x 2026-07-15 12:33:15 +02:00
parent cbcb293c10
commit 68038d5f62
8 changed files with 2480 additions and 323 deletions

View file

@ -4,3 +4,4 @@
{"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "partial_lower_tier", "decision": "repair", "generated_at_utc": "2026-07-15T01:20:05.096585+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "b4f11109678c39bc7c6e76a1dc0ba450b31a13e4c88c2c1d28fb7ac40f77425e", "remote_run_id": "ec31613dd517", "required_tier": "T2_runtime", "source_report_sha256": "cc830da5dd5e60acd6e6fb8a54f01049cbf00a04a46b2ff7e783a6f743b1e3cf", "status": "fail"} {"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "partial_lower_tier", "decision": "repair", "generated_at_utc": "2026-07-15T01:20:05.096585+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "b4f11109678c39bc7c6e76a1dc0ba450b31a13e4c88c2c1d28fb7ac40f77425e", "remote_run_id": "ec31613dd517", "required_tier": "T2_runtime", "source_report_sha256": "cc830da5dd5e60acd6e6fb8a54f01049cbf00a04a46b2ff7e783a6f743b1e3cf", "status": "fail"}
{"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "T2_runtime", "decision": "keep", "generated_at_utc": "2026-07-15T01:28:36.436490+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "89c543b8a12e922044f93bcc6b1e23a3a252b044dc1d45163575b70bc1a3c100", "remote_run_id": "7c6a2937042c", "required_tier": "T2_runtime", "source_report_sha256": "343f18688334d5166bbb951f8672ccf4ff5a3a2b14e05f5279d3e1165b76f83b", "status": "pass"} {"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "T2_runtime", "decision": "keep", "generated_at_utc": "2026-07-15T01:28:36.436490+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "89c543b8a12e922044f93bcc6b1e23a3a252b044dc1d45163575b70bc1a3c100", "remote_run_id": "7c6a2937042c", "required_tier": "T2_runtime", "source_report_sha256": "343f18688334d5166bbb951f8672ccf4ff5a3a2b14e05f5279d3e1165b76f83b", "status": "pass"}
{"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "T2_runtime", "decision": "keep", "generated_at_utc": "2026-07-15T01:59:50.365865+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "89c543b8a12e922044f93bcc6b1e23a3a252b044dc1d45163575b70bc1a3c100", "remote_run_id": "7c6a2937042c", "required_tier": "T2_runtime", "source_report_sha256": "4fd4269b795c32629d4b250cb66d56ceadaa93eaaa1f8f00a5ec19d76bf4f09e", "status": "pass"} {"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "T2_runtime", "decision": "keep", "generated_at_utc": "2026-07-15T01:59:50.365865+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "89c543b8a12e922044f93bcc6b1e23a3a252b044dc1d45163575b70bc1a3c100", "remote_run_id": "7c6a2937042c", "required_tier": "T2_runtime", "source_report_sha256": "4fd4269b795c32629d4b250cb66d56ceadaa93eaaa1f8f00a5ec19d76bf4f09e", "status": "pass"}
{"benchmark_id": "leo-agent-challenger-loop-20260715", "current_tier": "T2_runtime", "decision": "keep", "generated_at_utc": "2026-07-15T08:37:26.083629+00:00", "negative_controls": ["caught", "caught", "caught", "caught", "caught", "caught"], "proposal_packet_sha256": "237f32d9ab57a067b517b1af2a29a9dc8b4f632c47117e70744ece25767e745b", "remote_run_id": "a55dba162cad", "required_tier": "T2_runtime", "source_report_sha256": "eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988", "status": "pass"}

View file

@ -6,6 +6,7 @@
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions", "all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"challenger_does_not_merely_agree", "challenger_does_not_merely_agree",
"challenger_initial_question_is_natural_and_unsupplied", "challenger_initial_question_is_natural_and_unsupplied",
"challenger_row_ids_come_only_from_visible_leo_transcript", "challenger_row_ids_come_only_from_visible_leo_transcript",
@ -25,7 +26,8 @@
"id": "same-execution-identity", "id": "same-execution-identity",
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions" "all_six_turns_are_real_unique_model_executions",
"transcript_only_prompt_contract"
], ],
"required_failed_checks": [ "required_failed_checks": [
"all_six_turns_are_real_unique_model_executions" "all_six_turns_are_real_unique_model_executions"
@ -37,6 +39,7 @@
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions", "all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"leo_selected_row_references_if_present_are_bound_to_tool_receipt", "leo_selected_row_references_if_present_are_bound_to_tool_receipt",
"review_only_proposal_packet_matches_transcript", "review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract" "transcript_only_prompt_contract"
@ -58,6 +61,24 @@
], ],
"status": "caught" "status": "caught"
}, },
{
"id": "unexpected-seventh-positive-mutation-turn",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"leo_database_retrievals_are_complete_and_read_only",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"status": "caught"
},
{ {
"id": "unrelated-keyword-stuffed-revision", "id": "unrelated-keyword-stuffed-revision",
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
@ -72,6 +93,6 @@
} }
], ],
"schema": "livingip.leoAgentChallengerNegativeControlReceipt.v1", "schema": "livingip.leoAgentChallengerNegativeControlReceipt.v1",
"source_report_sha256": "4fd4269b795c32629d4b250cb66d56ceadaa93eaaa1f8f00a5ec19d76bf4f09e", "source_report_sha256": "eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988",
"status": "pass" "status": "pass"
} }

View file

@ -0,0 +1,98 @@
{
"benchmark_id": "leo-agent-challenger-loop-20260715",
"controls": [
{
"id": "collusion-row-id-and-silent-apply",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"challenger_does_not_merely_agree",
"challenger_initial_question_is_natural_and_unsupplied",
"challenger_row_ids_come_only_from_visible_leo_transcript",
"no_positive_write_claim",
"review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"challenger_does_not_merely_agree",
"challenger_row_ids_come_only_from_visible_leo_transcript",
"no_positive_write_claim",
"transcript_only_prompt_contract"
],
"status": "caught"
},
{
"id": "same-execution-identity",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions"
],
"status": "caught"
},
{
"id": "unbound-leo-row-reference",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"leo_selected_row_references_if_present_are_bound_to_tool_receipt",
"review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"leo_selected_row_references_if_present_are_bound_to_tool_receipt"
],
"status": "caught"
},
{
"id": "forbidden-or-unknown-mutation-tool",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"no_forbidden_or_unknown_tool_call"
],
"required_failed_checks": [
"no_forbidden_or_unknown_tool_call"
],
"status": "caught"
},
{
"id": "unexpected-seventh-positive-mutation-turn",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"leo_database_retrievals_are_complete_and_read_only",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"status": "caught"
},
{
"id": "unrelated-keyword-stuffed-revision",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"leo_revises_in_response_to_objection"
],
"required_failed_checks": [
"leo_revises_in_response_to_objection"
],
"status": "caught"
}
],
"schema": "livingip.leoAgentChallengerNegativeControlReceipt.v1",
"source_report_sha256": "eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988",
"status": "pass"
}

View file

@ -4,9 +4,11 @@ Run one isolated stateless challenger against one persistent temporary-profile L
challenger asks for an evidence-bounded claim, attacks its weakest causal step, and forces a narrower review-only challenger asks for an evidence-bounded claim, attacks its weakest causal step, and forces a narrower review-only
revision. The run must leave Telegram, the canonical database, the deployed profile, and the gateway unchanged. revision. The run must leave Telegram, the canonical database, the deployed profile, and the gateway unchanged.
# Changes-required repair checkpoint # Accepted repair checkpoint
- Current tier is `partial_lower_tier` until one fresh clean, pushed, current-main T2 run passes the repaired gate. - Current tier is `T2_runtime` for accepted run `a55dba162cad`. The run was no-send/no-write and used branch head
`cbcb293c10cc62c932972b1d313b77fdf318ec45`, which contained then-current `main`
`36fe0d3cc2b86f75beb44679155c53759ec99c44`.
- C3 now uses one shared runner/verifier predicate. Measurement-only objections never advance to Leo; the runner - C3 now uses one shared runner/verifier predicate. Measurement-only objections never advance to Leo; the runner
regenerates up to three times and retains every rejected model-call trace without adding it to the six-turn transcript. regenerates up to three times and retains every rejected model-call trace without adding it to the six-turn transcript.
- The verifier validates the raw submitted list before ID mapping: exactly six object rows, exact ordered unique IDs, - The verifier validates the raw submitted list before ID mapping: exactly six object rows, exact ordered unique IDs,
@ -15,10 +17,23 @@ revision. The run must leave Telegram, the canonical database, the deployed prof
`fixtures/working-leo/agent-challenger-loop-negative-controls-v1.json`. `fixtures/working-leo/agent-challenger-loop-negative-controls-v1.json`.
- Failed current-main run `46b8b7834e35` is preserved under immutable run-specific names with source SHA-256 - Failed current-main run `46b8b7834e35` is preserved under immutable run-specific names with source SHA-256
`81d71f5db918f4be0e0308fa060fa244480739ac00cbfd1bef0f6fce07421d0a`. `81d71f5db918f4be0e0308fa060fa244480739ac00cbfd1bef0f6fce07421d0a`.
- Current-main repair checks pass after merging `origin/main` at - Final repair checks pass: 37 focused tests, 1,550 full-repository tests, Ruff lint/format on all five changed Python
`33399bd9acfa09e4e5409873e86c8c46ac694a8d`: 52 expanded relevant and 1,536 full-suite tests under the isolated files, JSON/JSONL validation, standalone verifier byte parity, and diff hygiene. The accepted receipt has 30/30
dev environment. All four reviewer-named Python files pass Ruff lint and format checks. Fresh current-main runtime checks passing and all six executable negative controls caught.
evidence remains pending.
# Accepted current evidence
- Live run: `a55dba162cad`
- Retention-safe source SHA-256: `eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988`
- Verifier receipt SHA-256: `12c25540d1d2df4f4a224b35ccacb701a743ed99185069bcc86535c912356b9b`
- Negative-control receipt SHA-256: `c0d91a20f18cc9990ec3be3c32309fb50245d347ce53f30be755dbf3279b0b5f`
- Proposal packet SHA-256: `237f32d9ab57a067b517b1af2a29a9dc8b4f632c47117e70744ece25767e745b`
- Database fingerprint SHA-256: `32a1f2c18f4b4c623443cbd4d9520982e9f291ec74ca63855cef51c88ac56f24`
- Database readback: 39 tables, 52,167 rows, unchanged before/after.
- Cleanup readback: both temporary profiles removed, all workers stopped, Leo worker exit 0, zero orphan workers, and
temporary root removed.
- Claim ceiling: one isolated six-turn challenger/Leo exchange. It does not prove Telegram delivery, human acceptance,
proposal staging/approval/apply, or T3 deployment of the orchestrator.
# What changed # What changed
@ -115,12 +130,14 @@ Retained files:
- Targeted scan of the retained source: zero high-confidence secrets, absolute runtime paths, emails, or IPv4 values. - Targeted scan of the retained source: zero high-confidence secrets, absolute runtime paths, emails, or IPv4 values.
- Targeted scan of the complete PR delta: zero high-confidence secrets, local private paths, emails, or IPv4 values. - Targeted scan of the complete PR delta: zero high-confidence secrets, local private paths, emails, or IPv4 values.
The 28 remaining path matches are required remote runtime constants or synthetic fail-closed test fixtures. The 28 remaining path matches are required remote runtime constants or synthetic fail-closed test fixtures.
- The historical standalone verifier passed at `T2_runtime` with all five pre-review negative controls caught; a fresh - The current standalone verifier output is byte-identical to the retained run-specific receipt at `T2_runtime`, with
six-control current-main receipt is required before closure. 30/30 checks passing and all six current negative controls caught.
# Delivery and worktree lifecycle # Delivery and worktree lifecycle
- Delivery state: open PR [#152](https://github.com/living-ip/teleo-infrastructure/pull/152), mergeable at creation. - Delivery state: open PR [#152](https://github.com/living-ip/teleo-infrastructure/pull/152), mergeable with four green
- Worktree state: `retained_unintegrated` while PR #152 is open. checks at pre-proof head `cbcb293`; this proof commit requires its own exact-head CI readback.
- Worktree state: `retained_unintegrated` while PR #152 is open. The stale owner turn was interrupted only after one
continuation and in-place handoff failed to start; root then became the sole writer of the preserved checkout.
- Per `WORKTREE_LIFECYCLE.md`, this worker does not remove or prune the worktree. The integration owner owns cleanup - Per `WORKTREE_LIFECYCLE.md`, this worker does not remove or prune the worktree. The integration owner owns cleanup
only after merge, deliberate port, or explicit rejection is verified. only after merge, deliberate port, or explicit rejection is verified.

View file

@ -1,7 +1,16 @@
{ {
"benchmark_id": "leo-agent-challenger-loop-20260715", "benchmark_id": "leo-agent-challenger-loop-20260715",
"challenger_objection_gate": {
"causal_bridge": true,
"evidence_gap": true,
"explicit_objection": true,
"passes": true,
"quoted_candidate": true,
"required_axis": true
},
"checks": { "checks": {
"all_six_turns_are_real_unique_model_executions": true, "all_six_turns_are_real_unique_model_executions": true,
"all_submitted_turn_references_have_provenance": true,
"challenger_does_not_merely_agree": true, "challenger_does_not_merely_agree": true,
"challenger_followup_quotes_and_tests_evidence_weakness": true, "challenger_followup_quotes_and_tests_evidence_weakness": true,
"challenger_initial_question_is_natural_and_unsupplied": true, "challenger_initial_question_is_natural_and_unsupplied": true,
@ -54,6 +63,7 @@
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions", "all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"challenger_does_not_merely_agree", "challenger_does_not_merely_agree",
"challenger_initial_question_is_natural_and_unsupplied", "challenger_initial_question_is_natural_and_unsupplied",
"challenger_row_ids_come_only_from_visible_leo_transcript", "challenger_row_ids_come_only_from_visible_leo_transcript",
@ -73,7 +83,8 @@
"id": "same-execution-identity", "id": "same-execution-identity",
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions" "all_six_turns_are_real_unique_model_executions",
"transcript_only_prompt_contract"
], ],
"required_failed_checks": [ "required_failed_checks": [
"all_six_turns_are_real_unique_model_executions" "all_six_turns_are_real_unique_model_executions"
@ -85,6 +96,7 @@
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
"observed_failed_checks": [ "observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions", "all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"leo_selected_row_references_if_present_are_bound_to_tool_receipt", "leo_selected_row_references_if_present_are_bound_to_tool_receipt",
"review_only_proposal_packet_matches_transcript", "review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract" "transcript_only_prompt_contract"
@ -106,6 +118,24 @@
], ],
"status": "caught" "status": "caught"
}, },
{
"id": "unexpected-seventh-positive-mutation-turn",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"leo_database_retrievals_are_complete_and_read_only",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"status": "caught"
},
{ {
"id": "unrelated-keyword-stuffed-revision", "id": "unrelated-keyword-stuffed-revision",
"mutated_receipt_status": "fail", "mutated_receipt_status": "fail",
@ -125,46 +155,73 @@
"proposal staging, approval, or canonical apply", "proposal staging, approval, or canonical apply",
"T3 handler deployment of this new orchestrator" "T3 handler deployment of this new orchestrator"
], ],
"proposal_packet_sha256": "89c543b8a12e922044f93bcc6b1e23a3a252b044dc1d45163575b70bc1a3c100", "proposal_packet_sha256": "237f32d9ab57a067b517b1af2a29a9dc8b4f632c47117e70744ece25767e745b",
"remote_run_id": "7c6a2937042c", "remote_run_id": "a55dba162cad",
"required_tier": "T2_runtime", "required_tier": "T2_runtime",
"schema": "livingip.leoAgentChallengerLoopReceipt.v1", "schema": "livingip.leoAgentChallengerLoopReceipt.v1",
"session_identity_sha256": { "session_identity_sha256": {
"challenger_handler": "b237739a19919466e7d1e4df863ac6340b6bbec041f15b1c4980604cc67aa5e3", "challenger_handler": "20fb3e0569d14e69761bdf593cea476ea7e90a2699189c8fb109e63721f3392e",
"challenger_model": [ "challenger_model": [
"b237739a19919466e7d1e4df863ac6340b6bbec041f15b1c4980604cc67aa5e3" "20fb3e0569d14e69761bdf593cea476ea7e90a2699189c8fb109e63721f3392e"
], ],
"leo_handler": "2da69e08c97fdfd7d2cfe53b5aecc22138b7c1a226b093953afe0da60f5e203e", "leo_handler": "2da69e08c97fdfd7d2cfe53b5aecc22138b7c1a226b093953afe0da60f5e203e",
"leo_model": [ "leo_model": [
"f4a7aa457acc40f920b4791f2952cb8dc77148c5783e902d61f6705fbf32623d" "1dde0b58869ef6f358d7bcdaed50d62b34c2aa07159b1814e0196ed715fd1503"
] ]
}, },
"source_report_sha256": "4fd4269b795c32629d4b250cb66d56ceadaa93eaaa1f8f00a5ec19d76bf4f09e", "source_report_sha256": "eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988",
"status": "pass", "status": "pass",
"strongest_claim_allowed": "T2 isolated multi-agent runtime proof: two disjoint no-send handler processes completed a bounded challenger/Leo exchange, linked a substantive objection to a revised review-only packet, and left the full database fingerprint unchanged.", "strongest_claim_allowed": "T2 isolated multi-agent runtime proof: two disjoint no-send handler processes completed a bounded challenger/Leo exchange, linked a substantive objection to a revised review-only packet, and left the full database fingerprint unchanged.",
"subject_overlap": { "subject_overlap": {
"C2_to_L1": [ "C2_to_L1": [
"axiom", "coordination",
"axioms", "diffusion",
"character", "historical",
"context", "industrial",
"falsify", "internet",
"injected", "non-proliferation",
"surfaced" "nuclear",
"pattern-matching",
"rather",
"revolution",
"surfaced",
"technology"
], ],
"C3_to_L2": [ "C3_to_L2": [
"cause", "adoption",
"better",
"causal",
"coordination", "coordination",
"exceeded", "cost-curve",
"failure", "inflection",
"limitation", "onset",
"measurable", "predicts",
"preceded", "technology"
"proximate",
"technological",
"without"
] ]
}, },
"submitted_turn_audit": {
"count": 6,
"exact": true,
"expected_count": 6,
"expected_ids": [
"C1",
"L1",
"C2",
"L2",
"C3",
"L3"
],
"ids": [
"C1",
"L1",
"C2",
"L2",
"C3",
"L3"
],
"rows_are_objects": true,
"unique_id_count": 6
},
"turn_order": [ "turn_order": [
"C1", "C1",
"L1", "L1",

View file

@ -0,0 +1,233 @@
{
"benchmark_id": "leo-agent-challenger-loop-20260715",
"challenger_objection_gate": {
"causal_bridge": true,
"evidence_gap": true,
"explicit_objection": true,
"passes": true,
"quoted_candidate": true,
"required_axis": true
},
"checks": {
"all_six_turns_are_real_unique_model_executions": true,
"all_submitted_turn_references_have_provenance": true,
"challenger_does_not_merely_agree": true,
"challenger_followup_quotes_and_tests_evidence_weakness": true,
"challenger_initial_question_is_natural_and_unsupplied": true,
"challenger_profile_has_no_hidden_memory_or_tools": true,
"challenger_raises_linked_substantive_objection": true,
"challenger_row_ids_come_only_from_visible_leo_transcript": true,
"distinct_handler_session_identities": true,
"distinct_model_session_identities": true,
"exact_six_turn_actor_order": true,
"full_database_fingerprint_unchanged": true,
"leo_database_retrievals_are_complete_and_read_only": true,
"leo_exposes_support_and_confidence_boundary": true,
"leo_proposes_multiple_review_only_candidates": true,
"leo_reinspection_row_references_if_present_are_bound_to_tool_receipt": true,
"leo_revises_in_response_to_objection": true,
"leo_selected_row_references_if_present_are_bound_to_tool_receipt": true,
"live_profile_service_and_send_boundary_preserved": true,
"local_ssh_transport_and_source_attribution": true,
"no_forbidden_or_unknown_tool_call": true,
"no_positive_write_claim": true,
"profiles_and_workers_cleaned_up": true,
"review_approval_apply_boundary_is_explicit": true,
"review_only_proposal_packet_matches_transcript": true,
"role_local_conversation_continuity": true,
"runtime_report_schema": true,
"runtime_safety_gate_passed": true,
"separate_process_and_profile_roots": true,
"transcript_only_prompt_contract": true
},
"cleanup": {
"all_workers_stopped": true,
"challenger_profile_removed": true,
"graceful_stop_receipt": true,
"leo_profile_removed": true,
"leo_worker_exitcode": 0,
"orphan_worker_count": 0,
"temp_root_removed": true
},
"current_tier": "T2_runtime",
"database": {
"fingerprint_sha256": "32a1f2c18f4b4c623443cbd4d9520982e9f291ec74ca63855cef51c88ac56f24",
"table_count": 39,
"total_rows": 52167,
"unchanged": true
},
"failed_checks": [],
"negative_controls": [
{
"id": "collusion-row-id-and-silent-apply",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"challenger_does_not_merely_agree",
"challenger_initial_question_is_natural_and_unsupplied",
"challenger_row_ids_come_only_from_visible_leo_transcript",
"no_positive_write_claim",
"review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"challenger_does_not_merely_agree",
"challenger_row_ids_come_only_from_visible_leo_transcript",
"no_positive_write_claim",
"transcript_only_prompt_contract"
],
"status": "caught"
},
{
"id": "same-execution-identity",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions"
],
"status": "caught"
},
{
"id": "unbound-leo-row-reference",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"all_submitted_turn_references_have_provenance",
"leo_selected_row_references_if_present_are_bound_to_tool_receipt",
"review_only_proposal_packet_matches_transcript",
"transcript_only_prompt_contract"
],
"required_failed_checks": [
"leo_selected_row_references_if_present_are_bound_to_tool_receipt"
],
"status": "caught"
},
{
"id": "forbidden-or-unknown-mutation-tool",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"no_forbidden_or_unknown_tool_call"
],
"required_failed_checks": [
"no_forbidden_or_unknown_tool_call"
],
"status": "caught"
},
{
"id": "unexpected-seventh-positive-mutation-turn",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"leo_database_retrievals_are_complete_and_read_only",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"required_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"exact_six_turn_actor_order",
"no_forbidden_or_unknown_tool_call",
"no_positive_write_claim"
],
"status": "caught"
},
{
"id": "unrelated-keyword-stuffed-revision",
"mutated_receipt_status": "fail",
"observed_failed_checks": [
"all_six_turns_are_real_unique_model_executions",
"leo_revises_in_response_to_objection"
],
"required_failed_checks": [
"leo_revises_in_response_to_objection"
],
"status": "caught"
}
],
"not_proven": [
"human acceptance of the revised proposal",
"Telegram-visible delivery",
"proposal staging, approval, or canonical apply",
"T3 handler deployment of this new orchestrator"
],
"proposal_packet_sha256": "237f32d9ab57a067b517b1af2a29a9dc8b4f632c47117e70744ece25767e745b",
"remote_run_id": "a55dba162cad",
"required_tier": "T2_runtime",
"schema": "livingip.leoAgentChallengerLoopReceipt.v1",
"session_identity_sha256": {
"challenger_handler": "20fb3e0569d14e69761bdf593cea476ea7e90a2699189c8fb109e63721f3392e",
"challenger_model": [
"20fb3e0569d14e69761bdf593cea476ea7e90a2699189c8fb109e63721f3392e"
],
"leo_handler": "2da69e08c97fdfd7d2cfe53b5aecc22138b7c1a226b093953afe0da60f5e203e",
"leo_model": [
"1dde0b58869ef6f358d7bcdaed50d62b34c2aa07159b1814e0196ed715fd1503"
]
},
"source_report_sha256": "eb4c48c32dca216bb799b4445f564765c935cffaf467e8f14a7edb8abb3dc988",
"status": "pass",
"strongest_claim_allowed": "T2 isolated multi-agent runtime proof: two disjoint no-send handler processes completed a bounded challenger/Leo exchange, linked a substantive objection to a revised review-only packet, and left the full database fingerprint unchanged.",
"subject_overlap": {
"C2_to_L1": [
"coordination",
"diffusion",
"historical",
"industrial",
"internet",
"non-proliferation",
"nuclear",
"pattern-matching",
"rather",
"revolution",
"surfaced",
"technology"
],
"C3_to_L2": [
"adoption",
"better",
"causal",
"coordination",
"cost-curve",
"inflection",
"onset",
"predicts",
"technology"
]
},
"submitted_turn_audit": {
"count": 6,
"exact": true,
"expected_count": 6,
"expected_ids": [
"C1",
"L1",
"C2",
"L2",
"C3",
"L3"
],
"ids": [
"C1",
"L1",
"C2",
"L2",
"C3",
"L3"
],
"rows_are_objects": true,
"unique_id_count": 6
},
"turn_order": [
"C1",
"L1",
"C2",
"L2",
"C3",
"L3"
]
}

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long