types
diapason.learning.spec_search.diagnose.types
¶
Data types for the diagnose phase.
Lightweight dataclasses used as return types by diagnostic tools and as internal data carriers. These are NOT pydantic models — they don't need validation or JSON schema generation.
See spec §5.2 for the tool return type rationale.
BenchmarkTask
dataclass
¶
One task from the personal benchmark.
ComparisonResult
dataclass
¶
Structured comparison between student and teacher outputs.
DiagnosticTool
dataclass
¶
A tool exposed to the teacher in the diagnose phase.
Unlike BaseTool, these are not registered in ToolRegistry.
They are lightweight wrappers: a name, description, JSON schema
for parameters, and a callable that implements the tool.
to_openai_function()
¶
Convert to OpenAI function-calling format.
StudentRun
dataclass
¶
Result of re-executing the local student on a benchmark task.
TeacherRun
dataclass
¶
Result of the teacher running itself on a benchmark task.
ToolCallRecord
dataclass
¶
One teacher tool call, persisted to the JSONL log.
to_jsonl_dict()
¶
Serialize to a JSON-safe dict for JSONL output.
ToolMeta
dataclass
¶
Metadata about a tool in the ToolRegistry.
TraceMeta
dataclass
¶
Lightweight summary of a trace for browsing.