wire-frame-processor

name	wire-frame-processor
description	Wire a new pipecat FrameProcessor into the ORBIS pipeline — class authoring through pipeline insertion

/wire-frame-processor

When to use

Use this when adding a new processing stage to the ORBIS pipecat pipeline: writing the FrameProcessor subclass, wiring it into app.py, and positioning it correctly between existing pipeline components.

Steps

1. Author the processor class — Append the new FrameProcessor subclass to the appropriate module under agent/ (e.g. agent/intent.py). Include __init__, process_frame, and any supporting methods.
2. Refine to proper FrameProcessor — Ensure the class inherits from FrameProcessor, uses await self.push_frame(frame) for pass-through, and handles only the Frame types it cares about.
3. Update app.py imports — Add the new class to the import block at the top of app.py.
4. Construct the processor in run_bot — Instantiate the processor and any associated handlers/callbacks inside the run_bot function, near the other component constructions.
5. Insert into the pipeline — Place the processor in the Pipeline([...]) list at the correct position relative to existing components (e.g. between audio_tags and user_agg for pre-LLM interception).

Examples

• Adding IntentRouterProcessor between audio_tags and user_agg to intercept TranscriptionFrame before the LLM aggregator sees it.
• Adding a moderation FrameProcessor between user_agg and llm to filter or transform LLMMessagesFrame before generation.