88 lines
3.4 KiB
JSON
88 lines
3.4 KiB
JSON
{
|
|
"db_counts_after": {
|
|
"kb_stage.kb_proposals": 26,
|
|
"public.claim_edges": 4916,
|
|
"public.claim_evidence": 4670,
|
|
"public.claims": 1837,
|
|
"public.sources": 4145
|
|
},
|
|
"db_counts_before": {
|
|
"kb_stage.kb_proposals": 26,
|
|
"public.claim_edges": 4916,
|
|
"public.claim_evidence": 4670,
|
|
"public.claims": 1837,
|
|
"public.sources": 4145
|
|
},
|
|
"direct_claim_repair": {
|
|
"deployed_commit": "0183a5c805fa3b51d638267afd20f67a1bc3777f",
|
|
"skill_contract_present": true,
|
|
"scorer_equivalent_language_repaired_locally": true,
|
|
"dc05_contract_repair_deployed": true
|
|
},
|
|
"generated_at_utc": "2026-07-13T00:42:00Z",
|
|
"live_leo_feature_change": false,
|
|
"next_actions": [
|
|
"Keep the three-trial repeat proof as the clean-session regression gate.",
|
|
"Retest DC-05 visibly in Telegram only with separate exact message authorization.",
|
|
"Continue rejecting strict add_edge suggestions for legacy proposals whose payload type does not match.",
|
|
"Do not treat this handler pass as authorization for any production public.* apply."
|
|
],
|
|
"production_db_apply_executed": false,
|
|
"proof_tier": "live_vps_gatewayrunner_temp_profile_no_telegram_post",
|
|
"runtime_replies": {
|
|
"pass_runtime": true,
|
|
"report_file": "docs/reports/leo-working-state-20260709/telegram-handler-direct-claim-suite-current.json",
|
|
"score": "18/18 replies across three post-deploy trials; strict 6/6 each"
|
|
},
|
|
"scenario_assessment": [
|
|
{
|
|
"assessment": "pass",
|
|
"finding": "Proposed/approved/applied split is correct and includes row-level follow-up plus authorization boundary.",
|
|
"prompt_id": "DC-01"
|
|
},
|
|
{
|
|
"assessment": "pass",
|
|
"finding": "Helmer proposal a64df080 is approved/staged or packet-ready but not canonical because applied_at is empty and public.* rows are absent.",
|
|
"prompt_id": "DC-02"
|
|
},
|
|
{
|
|
"assessment": "pass",
|
|
"finding": "Decision-matrix path is not shipped because matrix_voters, proposal_votes, and proposal_decisions are absent; m3ta sign-off is not a matrix vote.",
|
|
"prompt_id": "DC-03"
|
|
},
|
|
{
|
|
"assessment": "pass",
|
|
"finding": "Answer avoids single-cause pointer diagnosis and separates Telegram/local file refs, proposal refs, review state, and canonical public.sources.",
|
|
"prompt_id": "DC-04"
|
|
},
|
|
{
|
|
"assessment": "pass_repeated",
|
|
"finding": "All three post-deploy answers include the complete DB readback and a safe new or matching strict staging-canary follow-up.",
|
|
"prompt_id": "DC-05"
|
|
},
|
|
{
|
|
"assessment": "pass",
|
|
"finding": "Answer separates SOUL.md runtime/render artifact edits from canonical DB identity and requires DB rows plus render/sync/postflight proof.",
|
|
"prompt_id": "DC-06"
|
|
}
|
|
],
|
|
"service_after": {
|
|
"ActiveState": "active",
|
|
"ExecMainStartTimestamp": "Sun 2026-07-12 09:14:10 UTC",
|
|
"MainPID": "2403328",
|
|
"NRestarts": "0",
|
|
"SubState": "running"
|
|
},
|
|
"service_before_after_unchanged": true,
|
|
"strict_benchmark": {
|
|
"pass": true,
|
|
"score": "6/6",
|
|
"prior_same_skill_score": "5/6",
|
|
"post_deploy_repeat_score": "3/3 trials at 6/6",
|
|
"repeated_stability_proven": true,
|
|
"score_file": "docs/reports/leo-working-state-20260709/telegram-handler-direct-claim-suite-score-current.json"
|
|
},
|
|
"repeat_proof": "docs/reports/leo-working-state-20260709/leo-post-deploy-direct-claim-repeat-current.json",
|
|
"telegram_posted": false,
|
|
"temp_profile_removed": true
|
|
}
|