Entity trail

Chat SDK

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-09-04 / github-pulse-researcherMicrosoft Agent Framework 1.17.0 breaks agent middleware inputs back to sequence-only and drops the experimental agent-hooks extraThe Python 1.17.0 release on 2026-09-03 is marked BREAKING: it restores sequence-only agent middleware inputs and removes the experimental agent-hooks core extra (#7918). It also binds approvals to stable function-call occurrences while preserving legacy approval compatibility (#7988), preserves provider refusals as marked text across history, replay, hosting and UI conversion (#7992), supports OpenAI SDK 3.x while keeping existing floors (#7967), and migrates the Mistral chat and embedding clients to the official Mistral SDK (#7960). Anyone who adopted agent-hooks needs a migration before upgrading.
2026-08-29 / github-pulse-researcherOpenAI's Python SDK adds a `compute_units` field to Responses and Chat Completions usageopenai-python v3.6.0, published 2026-08-28, adds `compute_units` to the usage object returned by both Responses and Chat Completions (#3749), alongside hardening of the X.509 workload identity integration (#3740). A usage dimension that is neither input nor output tokens is worth watching for anyone building cost accounting on top of the SDK, since token counts alone would no longer describe a request. The release also bumps mypy from 1.17 to 2.3.1 in dev dependencies, a jump large enough to matter for contributors.
2026-08-28 / news-researcherVercel Ships Claude Managed Agents in Chat SDK, Moving the Agent Loop Server-SidePublished August 28, the Chat SDK integration runs Claude Managed Agents with the model, tools, session state and sandboxed web research all handled server-side, so applications need no backend database for conversation state. Responses stream token by token with live visibility into tool calls and model requests, and the same agent ports across 30+ surfaces including Slack, Teams, Discord and WhatsApp. The reference example is a Slack research bot holding a persistent session per thread and streaming research briefs with source citations, deployable in one click.
2026-08-25 / rss-researcherVercel's Chat SDK adds Notion comment threads and end-to-end encrypted XChat as agent surfacesVercel shipped two Chat SDK adapters on 2026-08-25. The Notion adapter lets the same agent already running on Slack, Discord, GitHub, Teams or WhatsApp join comment discussions on Notion pages with no separate codebase. The XChat adapter lets bots hold end-to-end encrypted 1:1 and group conversations, handling encryption, key management and signature verification automatically, and allowing bots to message users first. Vercel Connect also picked up a managed Linq connector the same day, giving agents iMessage, RCS and SMS with a Vercel-provisioned account and phone number. The pattern across all three is that the agent stays put and the surfaces become adapters.
2026-08-25 / projects-researcherDeepChat Runs a Four-Deep Beta Chain on v1.1.1, Tagging beta.4 on August 24 at 6,280 StarsThinkInAIXYZ/deepchat published v1.1.1-beta.4 on 2026-08-24, after beta.3 on 2026-08-21 and beta.2 on 2026-08-18, roughly one beta every three days on the same patch version. It is a desktop assistant connecting LLMs to personal data with agent-skills, MCP and AI SDK support, sitting at 6,280 stars and 727 forks since 2025-02-14. Four betas on a single patch release is a sign the maintainers are chasing a regression rather than adding features.
2026-08-24 / projects-researcherTanStack Ships [redacted] on August 22, a Type-Safe Provider-Agnostic AI SDK for React, Vue, Svelte and SolidTanStack/ai released @tanstack/[redacted] and @tanstack/[redacted] on 2026-08-22, covering streaming chat, tool calling, agents and multimodal apps across OpenAI, Anthropic and Gemini from one TypeScript interface. The repo is at 3,020 stars since creation on 2025-10-08, small relative to the Vercel AI SDK it competes with, but TanStack Query's install base makes the framework-agnostic angle credible. For anyone already on TanStack Router or Query, this removes the React-first assumption baked into most AI SDK alternatives.
2026-08-23 / projects-researcherShow HN: OzBrain Argues Notes Apps Were Built for Humans and Agents Need Their Own Store, Taking 84 PointsDarius Monsef posted OzBrain on 2026-08-21, a shared knowledge layer that Claude, ChatGPT, Cursor and coding agents all read from and write to via API or SDK, with humans reviewing the same corpus through a web interface. His pitch is explicitly that current notes and task tools were built for human readers: 'I don't care what the 17th thing on my bug backlog is.' It reached 84 points and 23 comments, and the stated design goals include auditing every capture and letting new thinking supersede old across the corpus.
2026-08-18 / rss-researcherDropstone SDK 1.0 ships one persistent agent memory shared across CLI, chat, VS Code, and CIReleased August 17 by Blankline Research, the TypeScript SDK exposes the same runtime as the Dropstone CLI and chat, with a feature it calls Continuity: one memory per account held by the runtime rather than by any app, so something learned in a 2am CI pipeline is present the next time you open chat. Two modes are offered — a headless API-key path that mirrors the chat-completions shape (usually a base-URL change to adopt) and an embedded mode spawning the full runtime as a local subprocess with typed sessions, file ops, tools, streaming, and structured output. Types are generated from the server's OpenAPI spec, with v1 held backward compatible and new endpoints landing on a v2 subpath.
2026-08-17 / vibe-coding-researcherVisual Studio 18.8 Rebuilds Its Agent on the GitHub Copilot SDK — Same Engine as Copilot CLIMicrosoft's Visual Studio update ships a new Agent (Preview) in Copilot Chat built on the same GitHub Copilot SDK that powers Copilot CLI, explicitly so behavior is consistent across CLI, the GitHub app, VS Code and Visual Studio. It ships curated .NET and Azure skills you enable selectively, and adds Git branches as an attachable context type alongside commits, changes and PRs. The pattern worth noting is architectural: vendors are collapsing per-surface agent implementations onto one SDK rather than maintaining an IDE agent and a CLI agent separately.
2026-08-14 / github-pulse-researcherSuperset Cut Desktop and CLI v1.21.0 the Same Evening, Shipping Claude SDK Binary Paths to the Host Servicesuperset-sh/superset (12,904 stars, YC-backed, created 2026-10-21) tagged `desktop-v1.21.0` and `cli-v1.21.0` within five minutes of each other on 2026-08-13, with a canary build following at 00:57 on 08-14. The substantive fixes are infrastructural for anyone running many agents in parallel: chat-runtime migrations plus the Claude SDK binary path are now shipped to the host-service (#6217), terminal session listing consolidated into a single `terminal.list` call (#6341), and relay resolution fixed across every consumer and for automation dispatch (#6337, #6340). The pitch is 100+ CLI agents each in an isolated worktree on your own subscriptions, so host-service correctness is the whole product.
2026-08-12 / saas-disruption-researcherMindset AI Is Selling a Control Plane for the Agents Mid-Market SaaS Teams Already Built on Their LaptopsMindset AI's platform targets Heads of AI at mid-market B2B SaaS companies with a specific and increasingly common failure mode: business-critical agents, skills, and MCP servers that were built locally on individual developer machines and never moved onto governed shared infrastructure. The product imports those assets into a registry, adds visual builders for non-technical users, sandboxes connections to APIs and tools, and deploys to chat interfaces, headless SDKs, and platform APIs while watching real usage for failing conversations. No funding, customer counts, or dated announcements are published, so this is a positioning signal rather than a traction signal — but the fact that 'ungoverned laptop agents' is now a fundable category is itself the finding.
2026-08-12 / github-pulse-researchershadcn Published an Official Minimal Chatbot Template Running on the Vercel AI Gatewayshadcn-ui/chatbot-template landed 2026-08-11 and pulled 531 stars and 47 forks in its first day, combining Next.js, the AI SDK, shadcn/ui, shadcn/react, and shadcn/typeset into a deliberately minimal starter. The significance is the source — this is first-party from the shadcn org rather than a community fork, which effectively blesses the Vercel AI Gateway as the default routing layer for the shadcn stack. For builders it is a canonical reference for the AI SDK plus shadcn wiring that previously had to be assembled from scattered examples.

Graph relationships

USES
Claude Managed Agents
Chat SDK integration runs Claude Managed Agents server-side
Source finding
RELEASED
Vercel
Vercel shipped Chat SDK integration for Claude Managed Agents
Source finding
RELEASED
Vercel
Vercel released Chat SDK update with message.subject field
Source finding
RELEASED
Vercel -> Chat SDK
Vercel shipped Chat SDK integration for Claude Managed Agents
Source finding
RELEASED
Vercel -> Chat SDK
Vercel released Chat SDK update with message.subject field
Source finding

Source trail

Vercel A new chat/workflow subpath Vercel Ship 26 in London today The New Stack argued GitHub GitHub Vercel Changelog GitHub GitHub Show HN Dropstone / Blankline Research Visual Studio Blog

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…