# workflow_chat Service

This is a chat service for generating a workflow.yaml. It can chat with a user
and generate YAML when required.

## Usage

Send a POST request to `/services/workflow_chat` with a JSON body containing a
`content` and `history` field.

### CLI Call

```bash
 poetry run python services/entry.py workflow_chat tmp/input.json tmp/output.json
```

- make sure to replace `tmp/input.json` with the path to the input file and
  `tmp/output.json` with the path to the output file, or you can create tmp
  directory in the root of the project and run the command as is.

### Example Input

Simple input:

```json
{
  "content": "Whenever fridge statistics are send to you, parse and aggregate the data and upload to a collection in redis.",
  "existing_yaml": "",
  "history": [],
  "context": {},
  "stream": false,
  "read_only": false
}
```

**Request Parameters:**
- `content` (optional): The user's question or instruction
- `existing_yaml` (optional): Current workflow YAML to modify
- `errors` (optional): Error message to fix
- `history` (optional, default: `[]`): Array of previous conversation turns
- `context` (optional): Context including `page_name` for navigation tracking
- `stream` (optional, default: `false`): Enable streaming response
- `read_only` (optional, default: `false`): Enable read-only mode (IDs removed, no code preservation)
- `api_key` (optional): Anthropic API key (falls back to environment variable)

Second conversation turn example:

````json
{
  "content": "Actually I want to schedule it for midnight every day.",
  "existing_yaml": "name: fridge-statistics-processing\njobs:\n  parse-and-aggregate-fridge-data:\n    name: Parse and Aggregate Fridge Data\n    adaptor: '@openfn/language-common@latest'\n    body: '| // Add data parsing and aggregation operations here'\n  upload-to-redis:\n    name: Upload to Redis Collection\n    adaptor: '@openfn/language-redis@latest'\n    body: '| // Add Redis collection upload operations here'\ntriggers:\n  webhook:\n    type: webhook\n    enabled: false\nedges:\n  webhook->parse-and-aggregate-fridge-data:\n    source_trigger: webhook\n    target_job: parse-and-aggregate-fridge-data\n    condition_type: always\n    enabled: true\n  parse-and-aggregate-fridge-data->upload-to-redis:\n    source_job: parse-and-aggregate-fridge-data\n    target_job: upload-to-redis\n    condition_type: on_job_success\n    enabled: true\n",
  "history": [
    {
      "role": "user",
      "content": "Whenever fridge statistics are send to you, parse and aggregate the data and upload to a collection in redis."
    },
    {
      "role": "assistant",
      "content": "I'll create a workflow that processes fridge statistics through a webhook trigger, then aggregates and stores the data in Redis.\n\n```\nname: fridge-statistics-processing\njobs:\n  parse-and-aggregate-fridge-data:\n    name: Parse and Aggregate Fridge Data\n    adaptor: \"@openfn/language-common@latest\"\n    body: \"| // Add data parsing and aggregation operations here\"\n  upload-to-redis:\n    name: Upload to Redis Collection\n    adaptor: \"@openfn/language-redis@latest\"\n    body: \"| // Add Redis collection upload operations here\"\ntriggers:\n  webhook:\n    type: webhook\n    enabled: false\nedges:\n  webhook->parse-and-aggregate-fridge-data:\n    source_trigger: webhook\n    target_job: parse-and-aggregate-fridge-data\n    condition_type: always\n    enabled: true\n  parse-and-aggregate-fridge-data->upload-to-redis:\n    source_job: parse-and-aggregate-fridge-data\n    target_job: upload-to-redis\n    condition_type: on_job_success\n    enabled: true\n```"
    }
  ]
}
````

### Error input example

The service also takes an `errors` field, which triggers a modified prompt
targeted at identifying and fixing an error. If there is any text in `errors`,
the service will attempt to fix the YAML given in `existing_yaml`, or
alternatively, the conversation history, given the error message in `errors`. It
will return an answer whatever information is given, but it is best to give it
the `existing_yaml` and conversation `history`.

See the `fix_yaml_error_system_prompt` in `gen_project_prompts.yaml` to see the
exact instructions.

````json
{
  "existing_yaml": "name: fridge-statistics-processing\njobs:\n  parse-and-aggregate-fridge-data:\n    name: Parse and Aggregate Fridge Data\n    adaptor: '@openfn/language-commons@latest'\n    body: '| // Add data parsing and aggregation operations here'\n  upload-to-redis:\n    name: Upload to Redis Collection\n    adaptor: '@openfn/language-redis@latest'\n    body: '| // Add Redis collection upload operations here'\ntriggers:\n  webhook:\n    type: webhook\n    enabled: false\nedges:\n  webhook->parse-and-aggregate-fridge-data:\n    source_trigger: webhook\n    target_job: parse-and-aggregate-fridge-data\n    condition_type: always\n    enabled: true\n  parse-and-aggregate-fridge-data->upload-to-redis:\n    source_job: parse-and-aggregate-fridge-data\n    target_job: upload-to-redis\n    condition_type: on_job_success\n    enabled: true\n",
  "errors": "adaptor error",
  "history": [
    {
      "role": "user",
      "content": "Whenever fridge statistics are send to you, parse and aggregate the data and upload to a collection in redis."
    },
    {
      "role": "assistant",
      "content": "I'll create a workflow that processes fridge statistics through a webhook trigger, then aggregates and stores the data in Redis.\n\n```\nname: fridge-statistics-processing\njobs:\n  parse-and-aggregate-fridge-data:\n    name: Parse and Aggregate Fridge Data\n    adaptor: \"@openfn/language-common@latest\"\n    body: \"| // Add data parsing and aggregation operations here\"\n  upload-to-redis:\n    name: Upload to Redis Collection\n    adaptor: \"@openfn/language-redis@latest\"\n    body: \"| // Add Redis collection upload operations here\"\ntriggers:\n  webhook:\n    type: webhook\n    enabled: false\nedges:\n  webhook->parse-and-aggregate-fridge-data:\n    source_trigger: webhook\n    target_job: parse-and-aggregate-fridge-data\n    condition_type: always\n    enabled: true\n  parse-and-aggregate-fridge-data->upload-to-redis:\n    source_job: parse-and-aggregate-fridge-data\n    target_job: upload-to-redis\n    condition_type: on_job_success\n    enabled: true\n```"
    }
  ]
}
````

### Example Output

The service returns a JSON object with `response`, `response_yaml`, `history`, and `usage` fields. The `response_yaml` may be `null` if no YAML was generated.

**Response Fields:**
- `response`: The assistant's text response
- `response_yaml`: Generated workflow YAML (may be `null`)
- `history`: Updated conversation history with page prefixes (e.g., `[pg:workflow/my_workflow]`)
- `usage`: Token usage breakdown

````json
{
  "response": "This workflow will handle incoming fridge statistics via a webhook, process the data, and store it in Redis. I've chosen a webhook trigger since you're receiving data, and used the Redis adaptor for the final storage step.",
  "response_yaml": "name: Fridge-Statistics-Processing\njobs:\n  parse-fridge-data:\n    name: Parse and Aggregate Fridge Data\n    adaptor: '@openfn/language-common@latest'\n    body: '| // Add data parsing and aggregation operations here'\n  upload-to-redis:\n    name: Upload to Redis Collection\n    adaptor: '@openfn/language-redis@latest'\n    body: '| // Add Redis collection storage operations here'\ntriggers:\n  webhook:\n    type: webhook\n    enabled: false\nedges:\n  webhook->parse-fridge-data:\n    source_trigger: webhook\n    target_job: parse-fridge-data\n    condition_type: always\n    enabled: true\n  parse-fridge-data->upload-to-redis:\n    source_job: parse-fridge-data\n    target_job: upload-to-redis\n    condition_type: on_job_success\n    enabled: true\n",
  "history": [
    {
      "role": "user",
      "content": "[pg:workflow/fridge-statistics-processing] Whenever fridge statistics are send to you, parse and aggregate the data and upload to a collection in redis."
    },
    {
      "role": "assistant",
      "content": "This workflow will handle incoming fridge statistics via a webhook, process the data, and store it in Redis. I've chosen a webhook trigger since you're receiving data, and used the Redis adaptor for the final storage step."
    }
  ],
  "usage": {
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "input_tokens": 3250,
    "output_tokens": 281
  }
}
````

### Error handling

See above for using the service to correct a YAML given an error message (Error
input example).

Errors in the service output: The service could output invalid YAML or an
invalid OpenFn workflow.yaml. If the model output JSON or YAML are invalid, the
service will attempt to re-generate an answer (max once) before returning an
answer. If the validation fails twice, it could return invalid YAML in the
`yaml` output key. The tokens from the retry are summed before adding to the
`usage` key.

Invalid adaptors: Currently we log a warning if there are invalid adaptor names
in a valid YAML.