This is a comment left during a code review.
Path: src/kagan/server/requests.py
Line: 55

Comment:
**`agent_backend: str` rejects `null` from callers**

The old route used `payload.get("agent_backend", "")`, which treated an explicit `{"agent_backend": null}` JSON value as falsy and fell back to the settings default. Pydantic now rejects `null` for a bare `str` field with a 422 — a behavioral regression for any client that omits the key with a null rather than absent it.

To preserve the original semantics (absent _or_ null → fall back to settings), declare it as `str | None`:

```suggestion
    agent_backend: str | None = None
```

The existing fallback logic in `_routes.py` (`if not agent_backend`) already handles `None` correctly, so no route change is needed.

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: src/kagan/server/requests.py
Line: 76-81

Comment:
**Misleading error message for wrong-type input**

`_strip_text` is a `mode="before"` validator, so it runs before Pydantic's own type coercion. If a caller sends `{"text": 42}` (an integer), `not isinstance(v, str)` fires first and raises `ValueError("text is required")` — which is factually incorrect. The field was _provided_; it was just the wrong type.

Consider separating the two failure modes:

```suggestion
    @field_validator("text", mode="before")
    @classmethod
    def _strip_text(cls, v: object) -> str:
        if not isinstance(v, str):
            raise ValueError("text must be a string")
        if not v.strip():
            raise ValueError("text is required")
        return v.strip()
```

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: src/kagan/server/requests.py
Line: 21-47

Comment:
**Duplicated `_validate_criteria` validator**

`CreateTaskRequest` and `UpdateTaskRequest` define byte-for-byte identical `_validate_criteria` methods. A shared base class removes the duplication and makes future changes (e.g. verifying list items are actually strings) apply in one place:

```python
class _TaskFieldsMixin(BaseModel):
    acceptance_criteria: list[str] | None = None

    @field_validator("acceptance_criteria", mode="before")
    @classmethod
    def _validate_criteria(cls, v: object) -> list[str] | None:
        if v is None:
            return None
        if not isinstance(v, list):
            raise ValueError("acceptance_criteria must be a list of strings")
        return v

class CreateTaskRequest(_TaskFieldsMixin):
    title: str
    ...

class UpdateTaskRequest(_TaskFieldsMixin):
    title: str | None = None
    ...
```

How can I resolve this? If you propose a fix, please make it concise.