# Audita

Audita polishes transcript JSON using glossary-aware and LLM-backed correction
modules. It validates proposed corrections and produces corrected text plus
reports for review and subprocess integration.

## Quickstart

From a checkout, build with Go 1.24 or newer:

```sh
go build -o ./bin/audita ./cmd/audita
```

Processing requires a reachable [compatible LLM endpoint](docs/integrations/openai-compatible-llm.md)
and access to its configured model. Supply its credentials through the
[configuration mechanisms](docs/config.md#secrets); the
[minimal config](examples/minimal-config.yml) reads `AUDITA_LLM_API_KEY` and uses
the [default model and endpoint](docs/config.md#llm-settings). Set that variable
in your environment before using an endpoint that requires authentication.
For another endpoint or model, edit a copy of the config.

Run the included small input with the minimal config:

```sh
./bin/audita process examples/tiny-transcript.json \
  --config examples/minimal-config.yml \
  --glossary examples/tiny-glossary.yaml \
  --output corrected.json
```

This makes LLM requests. Inherited configuration environment overrides still
apply; see [precedence](docs/config.md#effective-precedence).

## Documentation

- [CLI reference](docs/cli.md)
- [Configuration and maintained examples](docs/config.md)
- [Operations and artifact lifecycle](docs/operations.md)
- [Transcript and glossary inputs](docs/integrations/transcript-glossary-files.md)
- [Output files and report formats](docs/integrations/output-files.md)
- [Subprocess integration](docs/integrations/subprocess.md)
- [LLM endpoint contract](docs/integrations/openai-compatible-llm.md)
- [Contributor guide](docs/development.md)