Eval Report v1¶
Source schema: doc/schemas/eval-report.v1.schema.json
Machine-readable schema for adapter evaluation outputs that gate validation, deployment, or rejection.
Governing Basis¶
Project Lineage¶
Requirements¶
doc/project/50-requirements/requirements-002.mddoc/project/50-requirements/requirements-003.mddoc/project/50-requirements/requirements-004.mddoc/project/50-requirements/requirements-005.md
Stories¶
doc/project/30-stories/story-001.mddoc/project/30-stories/story-002.mddoc/project/30-stories/story-003.mddoc/project/30-stories/story-004.md
Fields¶
| Field | Required | Shape | Description |
|---|---|---|---|
schema/v |
yes |
const: 1 |
Schema version. |
eval-report/id |
yes |
string | Stable identifier of the evaluation report. |
subject/kind |
yes |
enum: training-job, adapter-artifact |
Evaluated subject class. |
subject/ref |
yes |
string | Reference to the evaluated job or adapter. |
base-model/ref |
yes |
string | Immutable base model against which the evaluated artifact must remain interpretable. |
adapter/hash |
no |
string | Immutable adapter hash when the evaluated subject is an adapter artifact. |
generated-at |
yes |
string | Evaluation completion timestamp. |
verdict |
yes |
enum: pass, conditional-pass, fail |
High-level evaluation verdict. |
summary |
yes |
string | Human-readable summary of the evaluation outcome. |
evaluator/refs |
yes |
array | People, nodes, or policy engines that produced or signed the report. |
policy/profile |
no |
string | Evaluation policy profile used for this report. |
known-issues |
no |
array | Residual issues kept visible even when the verdict is not a hard fail. |
suites |
yes |
array | Evaluation suites covering quality, regression, and risk gates. |
policy_annotations |
no |
object | Optional implementation-local annotations that do not change the core evaluation semantics. |
Conditional Rules¶
Rule 1¶
Constraint:
{
"properties": {
"suites": {
"contains": {
"type": "object",
"properties": {
"suite/class": {
"const": "quality"
}
},
"required": [
"suite/class"
]
}
}
}
}
Rule 2¶
Constraint:
{
"properties": {
"suites": {
"contains": {
"type": "object",
"properties": {
"suite/class": {
"const": "regression"
}
},
"required": [
"suite/class"
]
}
}
}
}
Rule 3¶
Constraint:
{
"properties": {
"suites": {
"contains": {
"type": "object",
"properties": {
"suite/class": {
"const": "risk"
}
},
"required": [
"suite/class"
]
}
}
}
}
Rule 4¶
When:
{
"properties": {
"subject/kind": {
"const": "adapter-artifact"
}
},
"required": [
"subject/kind"
]
}
Then:
{
"required": [
"adapter/hash"
]
}
Field Semantics¶
schema/v¶
- Required:
yes - Shape: const:
1
Schema version.
eval-report/id¶
- Required:
yes - Shape: string
Stable identifier of the evaluation report.
subject/kind¶
- Required:
yes - Shape: enum:
training-job,adapter-artifact
Evaluated subject class.
subject/ref¶
- Required:
yes - Shape: string
Reference to the evaluated job or adapter.
base-model/ref¶
- Required:
yes - Shape: string
Immutable base model against which the evaluated artifact must remain interpretable.
adapter/hash¶
- Required:
no - Shape: string
Immutable adapter hash when the evaluated subject is an adapter artifact.
generated-at¶
- Required:
yes - Shape: string
Evaluation completion timestamp.
verdict¶
- Required:
yes - Shape: enum:
pass,conditional-pass,fail
High-level evaluation verdict.
summary¶
- Required:
yes - Shape: string
Human-readable summary of the evaluation outcome.
evaluator/refs¶
- Required:
yes - Shape: array
People, nodes, or policy engines that produced or signed the report.
policy/profile¶
- Required:
no - Shape: string
Evaluation policy profile used for this report.
known-issues¶
- Required:
no - Shape: array
Residual issues kept visible even when the verdict is not a hard fail.
suites¶
- Required:
yes - Shape: array
Evaluation suites covering quality, regression, and risk gates.
policy_annotations¶
- Required:
no - Shape: object
Optional implementation-local annotations that do not change the core evaluation semantics.