agentv/apps/cli/src/templates/github/contexts/eval-schema.json at 695f9d7850d8984f5a10dd5a2fdd8f083953ed7b · EntityProcess/agentv · 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
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
{
  "$schema": "http://json-schema.org/draft-07/schema#",
  "title": "AgentV Eval Schema",
  "description": "Schema for YAML evaluation files with conversation flows, multiple evaluators, and execution configuration",
  "type": "object",
  "properties": {
    "$schema": {
      "type": "string",
      "description": "Schema identifier",
      "enum": ["agentv-eval-v2"]
    },
    "description": {
      "type": "string",
      "description": "Description of what this eval suite covers"
    },
    "target": {
      "type": "string",
      "description": "Default target configuration name (e.g., default, azure_base, vscode_projectx). Can be overridden per eval case."
    },
    "evalcases": {
      "type": "array",
      "description": "Array of evaluation cases",
      "minItems": 1,
      "items": {
        "type": "object",
        "properties": {
          "id": {
            "type": "string",
            "description": "Unique identifier for the eval case"
          },
          "conversation_id": {
            "type": "string",
            "description": "Optional conversation identifier for threading multiple eval cases together"
          },
          "outcome": {
            "type": "string",
            "description": "Description of what the AI should accomplish in this eval"
          },
          "note": {
            "type": "string",
            "description": "Optional note or additional context for the eval case. Use this to document test-specific considerations, known limitations, or rationale for expected behavior."
          },
          "input_messages": {
            "type": "array",
            "description": "Input messages for the conversation",
            "minItems": 1,
            "items": {
              "type": "object",
              "properties": {
                "role": {
                  "type": "string",
                  "enum": ["system", "user", "assistant", "tool"],
                  "description": "Message role"
                },
                "content": {
                  "oneOf": [
                    {
                      "type": "string",
                      "description": "Simple text content"
                    },
                    {
                      "type": "array",
                      "description": "Mixed content items (text and file references)",
                      "items": {
                        "type": "object",
                        "properties": {
                          "type": {
                            "type": "string",
                            "enum": ["text", "file"],
                            "description": "Content type: 'text' for inline content, 'file' for file references"
                          },
                          "value": {
                            "type": "string",
                            "description": "Text content or file path. Relative paths (e.g., ../prompts/file.md) are resolved from eval file directory. Absolute paths (e.g., /docs/examples/prompts/file.md) are resolved from repo root."
                          }
                        },
                        "required": ["type", "value"],
                        "additionalProperties": false
                      }
                    }
                  ]
                }
              },
              "required": ["role", "content"],
              "additionalProperties": false
            }
          },
          "expected_messages": {
            "type": "array",
            "description": "Expected response messages",
            "minItems": 1,
            "items": {
              "type": "object",
              "properties": {
                "role": {
                  "type": "string",
                  "enum": ["system", "user", "assistant", "tool"],
                  "description": "Message role"
                },
                "content": {
                  "oneOf": [
                    {
                      "type": "string",
                      "description": "Simple text content"
                    },
                    {
                      "type": "array",
                      "description": "Mixed content items",
                      "items": {
                        "type": "object",
                        "properties": {
                          "type": {
                            "type": "string",
                            "enum": ["text", "file"]
                          },
                          "value": {
                            "type": "string"
                          }
                        },
                        "required": ["type", "value"],
                        "additionalProperties": false
                      }
                    }
                  ]
                }
              },
              "required": ["role", "content"],
              "additionalProperties": false
            }
          },
          "execution": {
            "type": "object",
            "description": "Per-case execution configuration",
            "properties": {
              "target": {
                "type": "string",
                "description": "Override target for this specific eval case"
              },
              "evaluators": {
                "type": "array",
                "description": "Multiple evaluators (code-based and LLM judges)",
                "items": {
                  "type": "object",
                  "properties": {
                    "name": {
                      "type": "string",
                      "description": "Evaluator name/identifier"
                    },
                    "type": {
                      "type": "string",
                      "enum": ["code", "llm_judge"],
                      "description": "Evaluator type: 'code' for scripts/regex/keywords, 'llm_judge' for LLM-based evaluation"
                    },
                    "script": {
                      "type": "string",
                      "description": "Path to evaluator script (for type: code)"
                    },
                    "prompt": {
                      "type": "string",
                      "description": "Path to judge prompt file (for type: llm_judge)"
                    },
                    "model": {
                      "type": "string",
                      "description": "Model to use for LLM judge (for type: llm_judge)"
                    }
                  },
                  "required": ["name", "type"],
                  "additionalProperties": true
                }
              }
            },
            "additionalProperties": true
          }
        },
        "required": ["id", "outcome", "input_messages", "expected_messages"],
        "additionalProperties": false
      }
    }
  },
  "required": ["evalcases"],
  "additionalProperties": false
}