Open-Workflow-Library/schemas/learning-event.schema.json at main · oxbshw/Open-Workflow-Library · 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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://open-workflow-library.dev/schemas/learning-event.schema.json",
  "title": "Learning Event",
  "description": "Evidence captured for the future human-reviewed self-improvement loop. A learning event records a failure, the root cause as best understood, and a proposed rule that could prevent the same class of failure. Learning events are NOT applied automatically. They are evidence; promotion into actual repair rules or wiki entries requires human review.",
  "type": "object",
  "required": [
    "eventId",
    "source",
    "failureType",
    "failureSummary",
    "humanReviewStatus",
    "createdAt"
  ],
  "additionalProperties": false,
  "properties": {
    "eventId": {
      "type": "string",
      "description": "Stable identifier for this learning event.",
      "minLength": 1
    },
    "source": {
      "type": "string",
      "description": "Where this event came from.",
      "enum": [
        "validation",
        "generation",
        "repair-attempt",
        "human-report",
        "ci",
        "unknown"
      ]
    },
    "workflowId": {
      "type": "string",
      "description": "Catalog ID of the workflow the event is associated with, if any."
    },
    "framework": {
      "type": "string",
      "enum": [
        "n8n",
        "dify",
        "langgraph",
        "node-red",
        "make",
        "zapier",
        "pipedream",
        "airflow",
        "temporal",
        "unknown"
      ]
    },
    "failureType": {
      "type": "string",
      "description": "Coarse category of failure, e.g. 'schema-mismatch', 'http-timeout', 'auth-error', 'llm-output-shape'."
    },
    "failureSummary": {
      "type": "string",
      "description": "Short human-readable description of what failed."
    },
    "rootCause": {
      "type": "string",
      "description": "Best-understood root cause. May be revised on review."
    },
    "successfulFix": {
      "type": "string",
      "description": "The change that resolved the failure, if known."
    },
    "proposedRule": {
      "type": "object",
      "description": "A candidate repair-rule or wiki entry that would prevent this failure in the future.",
      "properties": {
        "kind": {
          "type": "string",
          "enum": ["repair-rule", "wiki-pattern", "wiki-integration", "wiki-failure-case"]
        },
        "id": { "type": "string" },
        "body": { "type": "string" }
      },
      "additionalProperties": false
    },
    "confidence": {
      "type": "number",
      "minimum": 0,
      "maximum": 1,
      "description": "Estimated confidence that the root cause and proposed rule are correct. Calibration is best-effort."
    },
    "evidence": {
      "type": "array",
      "description": "References supporting the diagnosis: log excerpts (redacted), validation errors, related workflow paths, prior learning events.",
      "items": {
        "type": "object",
        "required": ["kind"],
        "properties": {
          "kind": {
            "type": "string",
            "enum": ["log", "validation-error", "workflow-ref", "learning-event-ref", "external-link"]
          },
          "ref": { "type": "string" },
          "excerpt": { "type": "string" }
        },
        "additionalProperties": false
      }
    },
    "humanReviewStatus": {
      "type": "string",
      "enum": ["pending", "approved", "rejected"]
    },
    "appliedToWiki": {
      "type": "boolean",
      "description": "True only if a maintainer promoted this event into a wiki entry or repair rule."
    },
    "createdAt": {
      "type": "string",
      "format": "date-time"
    }
  }
}