LiveKit Agents 1.7.0 (Python + JS) introduces PII redaction for Agent Observability
Observability privacy controls distinct from the expressive-mode blog row, covering Python and Node. Class-2: folded quiet patch(es) 1.7.1 into this row.
Details3 sources
livekit-agents and @livekit/agents 1.7.0 semantically redact detected entities from chat history and audio recordings and filter sensitive data from logs and traces; the Node core package lands on the same 1.7 line.
livekit-agents / @livekit/agents 1.7.0
Pipecat Context Hub 0.5.3 adds --framework-version latest
This is the groundwork for the release-tag default in 0.7.0 (2026-08-30-pipecat-context-hub-0-7-0, W35), which makes the same behaviour the default.
Details1 source
Pipecat Context Hub 0.5.3 lets refresh resolve its framework pin to the newest pipecat release tag on every run.
Pipecat Context Hub 0.5.3
Decagon launches Assist, a live AI copilot for human customer care teams
The release puts voice in scope alongside chat, email, and CRM workflows, but it is a representative-assist layer rather than a replacement for Decagon’s customer-facing agent.
Also Hear
Details2 sources · 1 X post
Decagon Assist gives human customer-care representatives shared context, workflow actions, drafted replies, and knowledge access inside the tools they already use, including live transcription and guidance on voice calls.
Decagon Assist
CekuraPlatformpartnership2026-08-19
Cekura and Deloitte Canada form an alliance for tested, observable voice agents
The pairing makes evaluation and production monitoring part of an SI-led deployment motion, with no named customer or deployment volume in the primary page.
Also Use
Details2 sources · 1 X post
Cekura and Deloitte Canada will pair Deloitte's agent design and deployment work with Cekura's testing and observability for Canadian enterprises.
Deloitte Canada alliance · delivered by Deloitte Canada
Most Canadian enterprises can build a voice agent faster than they can get approval to launch one.
Today we're announcing our alliance with @DeloitteCanada.
They design and deploy the agents. We prove they hold up before a customer hears them, and keep watching once they do…
https://x.com/cekuraAi/status/2090122866075505031
ElevenLabs opens an MCP control plane for voice and chat agents in Claude
Agent operations move from a vendor console into an MCP-compatible coding or assistant environment, while the ElevenLabs account remains the system being changed.
Also Use
Details2 sources · 1 X post
ElevenLabs made its hosted MCP available in Claude so teams can review performance, create agents, update configurations, and estimate language-model costs from an external AI client.
ElevenLabs MCP for Claude
Introducing the ElevenLabs MCP, now available in Claude.
Your team can manage your voice and chat agents where you already work - review recent performance, create new agents, update configurations, and even estimate LLM costs before changes go live.
https://x.com/ElevenLabs/status/2089353435397116280