# `Aludel.Evals.Report`
[🔗](https://github.com/ccarvalho-eng/aludel/blob/main/lib/aludel/evals/report.ex#L1)

A normalized, versioned representation of an evaluation suite run.

Reporters consume this struct instead of reading persistence schemas directly.
This keeps output formats stable while the database representation evolves.
Use `from_suite_run/1` to create a report and `to_map/1` when a plain map is
needed for serialization.

# `status`

```elixir
@type status() :: String.t()
```

The final quality decision: passed, failed, invalid, or unavailable.

# `t`

```elixir
@type t() :: %Aludel.Evals.Report{
  prompt_version_id: binary(),
  provider_id: binary(),
  quality_policy: map() | nil,
  results: [map()],
  schema_version: pos_integer(),
  status: status(),
  suite_id: binary(),
  suite_run_id: binary(),
  summary: map(),
  type: String.t()
}
```

# `from_suite_run`

```elixir
@spec from_suite_run(Aludel.Evals.SuiteRun.t()) :: t()
```

Builds a schema-version-2 report from a persisted suite run.

A quality-policy result determines the report status when present. Runs
without a policy pass only when they contain at least one result and every
result passed.

# `to_map`

```elixir
@spec to_map(t()) :: map()
```

Returns the report as a JSON-compatible map with string keys.

---

*Consult [api-reference.md](api-reference.md) for complete listing*