Reorganize Learning Pipecat and Fundamentals around the multi-agent path#859
Merged
Conversation
The UIWorker page lived in Fundamentals > Multi-agent, where this important feature was hard to discover. Move it into the Learning Pipecat sequence (before Agent Handoff), update inbound links, and add a redirect from the old path.
Surface the multi-agent learning material that was buried in Fundamentals: - Move UIWorker, Custom Voices per Agent, and the LLM-agent page into Learning Pipecat so the path runs: Multiple LLM Agents -> UIWorker -> Agent Handoff -> Custom Voices per Agent -> Job Coordination. - Rename the LLM-agent page to 'Multiple LLM Agents' (multiple-llm-agents) with a new intro framing separate agents that each own their own LLM. - Re-thread every affected 'What's next' card to follow the new order. - Rename the Fundamentals subgroup 'Multi-agent' to 'Multi-Agent Architecture' for the remaining internals pages. - Add redirects from all old paths and fix inbound links.
Lead with what the agent does (sees and drives the screen) rather than the voice-enabling angle. Slug stays ui-worker, so no path change or redirects; update the title, the card in Multiple LLM Agents, and the two API reference link-text references.
Surface the distributed and proxy-agent capabilities that were buried in Fundamentals by making them the scaling coda of the multi-agent path: - Move distributed-agents and proxy-agents into Learning Pipecat after Job Coordination (Job Coordination -> Distributed -> Proxy -> What's Next). - Re-thread the affected 'What's next' cards across Job Coordination, Distributed Agents, and Proxy Agents. - The remaining Multi-Agent Architecture subgroup is now purely the architectural primitives (bus, registry/discovery, bus bridge). - Add redirects from old paths, repoint /subagents redirects, and fix inbound cross-links.
Replace the roughly-alphabetical ordering with a thematic flow: conversation control & tuning, then capture & observe, then telephony patterns, then framework extension, with the Multi-Agent Architecture reference subgroup last. Kept as a flat list (plus the existing Evaluations and Multi-Agent Architecture subgroups); nav-only change.
Keep the three turn/input-control pages (muting, idle, stt-latency) contiguous, move context-summarization after them, and put saving-transcripts before recording-audio.
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
Contributor
Author
|
Contributor
|
This re-org makes sense. |
Contributor
Author
But it's not a fundamental either, I think. |
Contributor
|
Maybe just remove and let the example be the guide? |
The per-agent-TTS pattern is already shown in Multiple LLM Agents (the custom-pipeline section) and covered by an example, so the dedicated page is redundant. Remove it, re-thread Agent Handoff's 'What's next' to Job Coordination, repoint the old redirects to Multiple LLM Agents, and inline the custom-voices note there.
markbackman
approved these changes
Jun 1, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Surfaces Pipecat's multi-agent features (which were buried in Fundamentals) by promoting the full usage path into Learning Pipecat, and reorganizes the remaining Fundamentals section thematically.
Learning Pipecat: a complete multi-agent arc
The learning path now runs end to end through building and scaling a multi-agent system:
llm-agent-and-tool-decorator→multiple-llm-agents("Multiple LLM Agents") with a new intro framing separate agents that each own their own LLM.ui-worker), leading with what the agent does rather than the voice-enabling angle.Fundamentals
Multi-agentsubgroup →Multi-Agent Architecture— after the move it holds exactly the architectural primitives (bus, registry/discovery, bus bridge), and it stays at the bottom as optional reference.Plumbing
/pipecat/fundamentals/...path, repointed the/subagents/...redirects, and fixed all inbound links (API reference, getting-started, cross-links).docs.jsonvalidated;mint broken-linkspasses clean.🤖 Generated with Claude Code