NodeRoom/docs/eval/managed-lock-performance.json at main · HomenShum/NodeRoom · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
{
  "generatedAt": "2026-06-11T20:40:15.018Z",
  "mode": "deterministic_runtime",
  "route": "scripted",
  "caseId": "managed-lock-production-target-v1",
  "explicit": {
    "lane": "explicit_agent_lock",
    "status": "passed",
    "elapsedMs": 16,
    "modelName": "scripted-explicit-lock",
    "modelCalls": 7,
    "agentToolCalls": 6,
    "coordinationToolCalls": 2,
    "managedBatchCalls": 0,
    "traceTools": [
      "propose_lock",
      "read_range",
      "edit_cell",
      "read_range",
      "edit_cell",
      "release_lock"
    ],
    "valuesOk": true,
    "released": true,
    "exhausted": false,
    "stopReason": "done",
    "usage": {
      "inputTokens": 0,
      "outputTokens": 0,
      "modelCalls": 7
    },
    "coordinationEvidence": []
  },
  "managed": {
    "lane": "runtime_managed_lock",
    "status": "passed",
    "elapsedMs": 4,
    "modelName": "scripted-managed-lock",
    "modelCalls": 3,
    "agentToolCalls": 2,
    "coordinationToolCalls": 0,
    "managedBatchCalls": 1,
    "traceTools": [
      "read_range",
      "write_locked_cells"
    ],
    "valuesOk": true,
    "released": true,
    "exhausted": false,
    "stopReason": "done",
    "usage": {
      "inputTokens": 0,
      "outputTokens": 0,
      "modelCalls": 3
    },
    "coordinationEvidence": [
      {
        "mode": "managed_lock_batch",
        "targetIds": [
          "r_rev__variance",
          "r_cogs__variance"
        ],
        "acquired": true,
        "lockId": "lock_21",
        "released": true,
        "mergedDrafts": 0
      }
    ]
  },
  "delta": {
    "modelCallsSaved": 4,
    "agentToolCallsSaved": 4,
    "coordinationToolCallsRemovedFromModelTrace": 2,
    "agentToolCallReductionPct": 66.7
  },
  "lesson": {
    "giveAgent": [
      "business intent",
      "target cells",
      "values/formulas/evidence",
      "base versions from reads"
    ],
    "takeAwayFromAgent": [
      "lock acquisition",
      "unlock sequencing",
      "range coordination",
      "draft-on-blocked mechanics",
      "release-in-finally cleanup"
    ],
    "invariant": "The model-visible trace has fewer coordination calls, while the runtime result still carries lock coordination evidence and the room ends with no active locks."
  }
}