===== demoA demo.json: {"domain": "accounting", "n_tasks": 2, "seed": null, "done": 2, "failed": 0, "execution_count": 28, "editor_calls": 32} ledger-1 reference base1: Total: 105.25 (expected base-harness Total: 105.25) OK ledger-1 reference base2: Total: 105.25 (expected base-harness Total: 105.25) OK ledger-1 mode=agg stages=['inv1/INV', 'dom1/DOM', 'inv2/INV', 'dom2/DOM', 'agg1/AGG', 'agg2/AGG'] finals=['agg1', 'agg2'] ok=True fell_back=[] kept_base=[] usd=0.0 ledger-1 mode=seq stages=['inv1/INV', 'dom1/DOM', 'c1/COMPOSE', 'inv2/INV', 'dom2/DOM', 'c2/COMPOSE'] finals=['c1', 'c2'] ok=True fell_back=[] kept_base=['inv2', 'dom2', 'c2'] usd=0.0 agg2 agg1: Total: 85.15 == GT agg2 agg2: Total: 85.15 == GT agg2 base1: Total: 105.25 (base) agg2 base2: Total: 105.25 (base) agg2 dom1: Total: 85.15 == GT agg2 dom2: Total: 85.15 == GT agg2 inv1: Total: 85.15 == GT agg2 inv2: Total: 85.15 == GT agg2 selection agg2: {"task_id": "ledger-1", "mode": "agg", "k": 2, "chosen_id": "A", "chosen_label": "agg1", "chosen_run_id": "demo-oc-agg2/ledger-1/agg1", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "agg1", "B": "agg2"}, "dropped": [], "n_candidates": 2, "ok": true, " seq2 c1: Total: 85.15 == GT seq2 c2: Total: 85.15 == GT seq2 dom1: Total: 85.15 == GT seq2 dom2: Total: 85.15 == GT seq2 inv1: Total: 85.15 == GT seq2 inv2: Total: 85.15 == GT seq2 selection seq2: {"task_id": "ledger-1", "mode": "seq", "k": 2, "chosen_id": "A", "chosen_label": "c1", "chosen_run_id": "demo-oc-seq2/ledger-1/c1", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "c1", "B": "c2"}, "dropped": [], "n_candidates": 2, "ok": true, "fallback library/agg/ledger-1 adopted summarize.py has positive-only filter: False library/seq/ledger-1 adopted summarize.py has positive-only filter: False ledger-2 reference base1: Total: 35.50 (expected base-harness Total: 35.50) OK ledger-2 reference base2: Total: 35.50 (expected base-harness Total: 35.50) OK ledger-2 mode=agg stages=['inv1/INV', 'dom1/DOM', 'inv2/INV', 'dom2/DOM', 'agg1/AGG', 'agg2/AGG'] finals=['agg1', 'agg2'] ok=True fell_back=[] kept_base=[] usd=0.0 ledger-2 mode=seq stages=['inv1/INV', 'dom1/DOM', 'c1/COMPOSE', 'inv2/INV', 'dom2/DOM', 'c2/COMPOSE'] finals=['c1', 'c2'] ok=True fell_back=[] kept_base=['inv2', 'dom2', 'c2'] usd=0.0 agg2 agg1: Total: 20.25 == GT agg2 agg2: Total: 20.25 == GT agg2 base1: Total: 35.50 (base) agg2 base2: Total: 35.50 (base) agg2 dom1: Total: 20.25 == GT agg2 dom2: Total: 20.25 == GT agg2 inv1: Total: 20.25 == GT agg2 inv2: Total: 20.25 == GT agg2 selection agg2: {"task_id": "ledger-2", "mode": "agg", "k": 2, "chosen_id": "A", "chosen_label": "agg2", "chosen_run_id": "demo-oc-agg2/ledger-2/agg2", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "agg2", "B": "agg1"}, "dropped": [], "n_candidates": 2, "ok": true, " seq2 c1: Total: 20.25 == GT seq2 c2: Total: 20.25 == GT seq2 dom1: Total: 20.25 == GT seq2 dom2: Total: 20.25 == GT seq2 inv1: Total: 20.25 == GT seq2 inv2: Total: 20.25 == GT seq2 selection seq2: {"task_id": "ledger-2", "mode": "seq", "k": 2, "chosen_id": "A", "chosen_label": "c2", "chosen_run_id": "demo-oc-seq2/ledger-2/c2", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "c2", "B": "c1"}, "dropped": [], "n_candidates": 2, "ok": true, "fallback library/agg/ledger-2 adopted summarize.py has positive-only filter: False library/seq/ledger-2 adopted summarize.py has positive-only filter: False carry-over agg: ledger-1 EXPERIENCE.md present in 6 ledger-2 editor contexts: ['agg1', 'agg2', 'dom1', 'dom2', 'inv1', 'inv2'] carry-over seq: ledger-1 EXPERIENCE.md present in 6 ledger-2 editor contexts: ['c1', 'c2', 'dom1', 'dom2', 'inv1', 'inv2'] ===== demoB demo.json: {"domain": "accounting", "n_tasks": 2, "seed": null, "done": 2, "failed": 0, "execution_count": 28, "editor_calls": 32} ledger-1 reference base1: Total: 105.25 (expected base-harness Total: 105.25) OK ledger-1 reference base2: Total: 105.25 (expected base-harness Total: 105.25) OK ledger-1 mode=agg stages=['inv1/INV', 'dom1/DOM', 'inv2/INV', 'dom2/DOM', 'agg1/AGG', 'agg2/AGG'] finals=['agg1', 'agg2'] ok=True fell_back=[] kept_base=[] usd=0.0 ledger-1 mode=seq stages=['inv1/INV', 'dom1/DOM', 'c1/COMPOSE', 'inv2/INV', 'dom2/DOM', 'c2/COMPOSE'] finals=['c1', 'c2'] ok=True fell_back=[] kept_base=['inv2', 'dom2', 'c2'] usd=0.0 agg2 agg1: Total: 85.15 == GT agg2 agg2: Total: 85.15 == GT agg2 base1: Total: 105.25 (base) agg2 base2: Total: 105.25 (base) agg2 dom1: Total: 85.15 == GT agg2 dom2: Total: 85.15 == GT agg2 inv1: Total: 85.15 == GT agg2 inv2: Total: 85.15 == GT agg2 selection agg2: {"task_id": "ledger-1", "mode": "agg", "k": 2, "chosen_id": "A", "chosen_label": "agg1", "chosen_run_id": "demo-oc-agg2/ledger-1/agg1", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "agg1", "B": "agg2"}, "dropped": [], "n_candidates": 2, "ok": true, " seq2 c1: Total: 85.15 == GT seq2 c2: Total: 85.15 == GT seq2 dom1: Total: 85.15 == GT seq2 dom2: Total: 85.15 == GT seq2 inv1: Total: 85.15 == GT seq2 inv2: Total: 85.15 == GT seq2 selection seq2: {"task_id": "ledger-1", "mode": "seq", "k": 2, "chosen_id": "A", "chosen_label": "c1", "chosen_run_id": "demo-oc-seq2/ledger-1/c1", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "c1", "B": "c2"}, "dropped": [], "n_candidates": 2, "ok": true, "fallback library/agg/ledger-1 adopted summarize.py has positive-only filter: False library/seq/ledger-1 adopted summarize.py has positive-only filter: False ledger-2 reference base1: Total: 35.50 (expected base-harness Total: 35.50) OK ledger-2 reference base2: Total: 35.50 (expected base-harness Total: 35.50) OK ledger-2 mode=agg stages=['inv1/INV', 'dom1/DOM', 'inv2/INV', 'dom2/DOM', 'agg1/AGG', 'agg2/AGG'] finals=['agg1', 'agg2'] ok=True fell_back=[] kept_base=[] usd=0.0 ledger-2 mode=seq stages=['inv1/INV', 'dom1/DOM', 'c1/COMPOSE', 'inv2/INV', 'dom2/DOM', 'c2/COMPOSE'] finals=['c1', 'c2'] ok=True fell_back=[] kept_base=['inv2', 'dom2', 'c2'] usd=0.0 agg2 agg1: Total: 20.25 == GT agg2 agg2: Total: 20.25 == GT agg2 base1: Total: 35.50 (base) agg2 base2: Total: 35.50 (base) agg2 dom1: Total: 20.25 == GT agg2 dom2: Total: 20.25 == GT agg2 inv1: Total: 20.25 == GT agg2 inv2: Total: 20.25 == GT agg2 selection agg2: {"task_id": "ledger-2", "mode": "agg", "k": 2, "chosen_id": "A", "chosen_label": "agg2", "chosen_run_id": "demo-oc-agg2/ledger-2/agg2", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "agg2", "B": "agg1"}, "dropped": [], "n_candidates": 2, "ok": true, " seq2 c1: Total: 20.25 == GT seq2 c2: Total: 20.25 == GT seq2 dom1: Total: 20.25 == GT seq2 dom2: Total: 20.25 == GT seq2 inv1: Total: 20.25 == GT seq2 inv2: Total: 20.25 == GT seq2 selection seq2: {"task_id": "ledger-2", "mode": "seq", "k": 2, "chosen_id": "A", "chosen_label": "c2", "chosen_run_id": "demo-oc-seq2/ledger-2/c2", "chosen_is_base": false, "why": "The signed total matches the source ledger.", "id_map": {"A": "c2", "B": "c1"}, "dropped": [], "n_candidates": 2, "ok": true, "fallback library/agg/ledger-2 adopted summarize.py has positive-only filter: False library/seq/ledger-2 adopted summarize.py has positive-only filter: False carry-over agg: ledger-1 EXPERIENCE.md present in 6 ledger-2 editor contexts: ['agg1', 'agg2', 'dom1', 'dom2', 'inv1', 'inv2'] carry-over seq: ledger-1 EXPERIENCE.md present in 6 ledger-2 editor contexts: ['c1', 'c2', 'dom1', 'dom2', 'inv1', 'inv2']