The AI Toolchain — issue -070, June 10, 2026

Agno v2.6.13 adds sub-agent event streaming, AgentOS registry auto-population, socket-based HITL workflows, and a Slack app manifest.

$ git clone --branch v2.6.13 https://github.com/agno-agi/agno.git
# already have the repo? check out this version:
$ git checkout v2.6.13

• ›Sub-agent events from the context provider update tool now stream through to the parent run, enabling real-time observability of nested agent activity.
• ›The AgentOS registry now auto-populates from agents, teams, and workflows, eliminating manual registration.
• ›Adds socket support for human-in-the-loop (HITL) workflows, enabling interactive pause-and-resume over persistent socket connections.
• ›Adds a Slack app manifest for the AgentOS interface, simplifying Slack app setup and deployment.

AutoGPT v0.6.63 adds MCP auto-discovery, local-LLM AutoPilot, global search, file inputs for Replicate, and bot analytics.

$ git clone --branch autogpt-platform-beta-v0.6.63 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:
$ git checkout autogpt-platform-beta-v0.6.63

• ›Adds an ARTIFACTS_PAGE feature flag that gates a new Artifacts page in the frontend.
• ›Adds file input support to the Replicate model block, enabling image and file passing to Replicate-hosted models.
• ›Adds variable inputs to the Execute Code block for dynamic parameterization at runtime.
• ›Adds a global search endpoint plus a bucketed command-palette modal with navigate and action sections for fast in-app navigation.
• ›Adds bot usage analytics (events and presence) to the backend, surfaced through a new bot analytics admin page and read API.

LangChain 1.3.7 adds ProviderToolSearchMiddleware for filtering tool search by provider.

$ git clone --branch langchain==1.3.7 https://github.com/langchain-ai/langchain.git
# already have the repo? check out this version:
$ git checkout langchain==1.3.7

• ›Adds ProviderToolSearchMiddleware to enable provider-based filtering of tool search.

$ git clone --branch langchain-groq==1.1.3 https://github.com/langchain-ai/langchain.git
# already have the repo? check out this version:
$ git checkout langchain-groq==1.1.3

$ git clone --branch langchain==1.3.5 https://github.com/langchain-ai/langchain.git
# already have the repo? check out this version:
$ git checkout langchain==1.3.5

PydanticAI v1.107.0 adds known_model_names(), OpenRouter prompt caching, and two new Claude model aliases.

$ git clone --branch v1.107.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.107.0

from pydantic_ai import known_model_names

for name in known_model_names():
    print(name)

• ›Adds known_model_names() function to programmatically enumerate all KnownModelName members at runtime.
• ›Adds CachePoint and prompt caching support for OpenRouter models.
• ›Adds claude-fable-5 and claude-mythos-5 as supported model name aliases.

Adds Claude 4.5 ('claude fable 5') support and Anthropic extended thinking compatibility.

$ git clone --branch 0.13.1 https://github.com/browser-use/browser-use.git
# already have the repo? check out this version:
$ git checkout 0.13.1

• ›Supports Claude 4.5 (claude fable 5) as a model option for browser agents.
• ›Switches Anthropic tool choice to auto when extended thinking is enabled, enabling compatibility between thinking mode and tool use.

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Claude Code v2.1.172 adds recursive sub-agents (5 levels deep), Bedrock region auto-detection, plugin search, and a new OTEL model attribute.

$ git clone --branch v2.1.172 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.172

$ /status

• ›Enables sub-agents to spawn their own sub-agents, up to 5 levels of nesting.
• ›Amazon Bedrock now auto-detects AWS region from ~/.aws config files following AWS SDK precedence when AWS_REGION is unset; /status shows the region source.
• ›Adds a search bar inside the /plugin marketplace browser for filtering plugins.
• ›Adds a model attribute to the claude_code.lines_of_code.count OTEL metric, enabling per-model telemetry breakdowns.

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Cline v3.0.23 lets configured agents be used directly as subagent tools.

$ git clone --branch cli-v3.0.23 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout cli-v3.0.23

• ›Supports configured agents as subagent tools, enabling agent-to-agent delegation within Cline workflows.
• ›Centralizes OAuth management into the SDK, providing unified auth handling across integrations.

The open source coding agent.

OpenCode v1.17.1 adds richer reference metadata and a new desktop /new-session route with draft tabs.

$ git clone --branch v1.17.1 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.1

• ›References now support usage descriptions for agents, visibility control to hide entries from @ autocomplete, and inclusion in generated docs.
• ›New /new-session desktop route introduces draft tabs that keep prompt and project selection together until the session is started.

$ git clone --branch v1.17.0 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.0

OpenHands: AI-Driven Development

OpenHands 1.8.0 adds LLM profiles, sub-agent delegation, sandbox grouping strategies, and a generic ACP agent UI.

$ git clone --branch 1.8.0 https://github.com/All-Hands-AI/OpenHands.git
# already have the repo? check out this version:
$ git checkout 1.8.0

• ›Adds LLM profiles, letting users save and switch between named LLM configurations.
• ›Enables sub-agent delegation, allowing the primary agent to hand off subtasks to specialised child agents.
• ›Adds user-selectable sandbox grouping strategy to control how sandbox environments are allocated and shared.
• ›Introduces a minimal generic ACP (Agent Communication Protocol) agent UI for interacting with ACP-compatible agents.

Zed v1.6.3 adds agent 'Fast mode' for Anthropic/OpenAI, shareable skill links, Git split-diff history, and a new agent.commit_message_instructions setting.

$ git clone --branch v1.6.3 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v1.6.3

{
  "agent": {
    "commit_message_instructions": "Use Conventional Commits format: type(scope): short summary. Include a body only when the change is non-trivial."
  }
}

{
  "scan_symlinks": "always"
}

{
  "language_models": {
    "openai": {
      "custom_headers": {
        "X-Org-Id": "<your-org-id>"
      }
    }
  }
}

• ›Adds agent.commit_message_instructions setting to customize AI-generated git commit messages.
• ›Adds scan_symlinks setting to control symlinked directory scanning; accepts expanded (default, scans only after expansion in project panel) or always (follows all symlinks during indexing).
• ›Adds language_models.<provider>.custom_headers config key to set custom HTTP headers on language model providers.
• ›Adds new git: compare with branch action to directly compare the current branch with an arbitrary branch.
• ›Adds editor: toggle breadcrumb action to toggle the breadcrumb of the active item without persisting the change to settings.json.

LocalAI v4.4.0 adds two ASR backends, object detection, video I/O, distributed prefix-cache routing, PII middleware, and a local-ai chat CLI.

$ git clone --branch v4.4.0 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:
$ git checkout v4.4.0

$ curl http://localhost:8080/v1/audio/speech -H "Content-Type: application/json" -d '{"model": "qwen-tts-design", "input": "Hello world, this is a test.", "instructions": "A calm, low-pitched elderly storyteller with a warm tone."}'

• ›Adds LOCALAI_FILE_TRANSFER_BUDGET environment variable (default 1h) to control the time budget for resumable large-model GGUF transfers over the distributed worker PUT /v1/files/<key> endpoint, with Content-Range / 308/416 resume semantics and X-Content-SHA256 binding.
• ›Adds --distributed-prefix-cache=false flag to disable the new prefix-cache-aware routing in distributed mode (on by default), which biases requests toward the replica already holding the relevant KV/prefix cache.
• ›Adds LOCALAI_PREFETCH_MODELS environment variable for boot-time gallery model prefetch in distributed deployments.
• ›Plumbs instructions string and params map through the /v1/audio/speech request body end-to-end (proto, schema, core/backend/tts.go), enabling per-request voice/style control for Qwen3-TTS VoiceDesign, Qwen3-TTS CustomVoice, and Chatterbox.
• ›New crispasr backend (purego, CGO_ENABLED=0) serving ASR or TTS from a single binary, auto-detecting architecture from GGUF; supports backend:, codec:, speaker:, and voice: model options, with 36 gallery entries (32 ASR + 4 TTS including vibevoice, chatterbox, qwen3-tts, orpheus).

Phoenix v17.3.0 adds subagents, playground run controls, claude-fable-5 support, and expanded PXI agent tooling.

$ git clone --branch arize-phoenix-v17.3.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v17.3.0

• ›Adds set_appended_messages_path PXI tool for configuring appended messages paths in the playground.
• ›Adds playground prompt instance tools via the pxi agent toolset, enabling agent-driven prompt management.
• ›Adds playground experiment recording tool for capturing experiment results from within the PXI agent.
• ›Adds dataset management tools for the PXI agent, enabling agent-driven dataset operations.
• ›Adds a template variables path tool in the playground for setting template variable sources.

Langfuse v3.183.0 adds web callouts from traces and sessions, new no-data resolution modes for monitors, and user location context in the agent.

$ git clone --branch v3.183.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.183.0

• ›Adds additional no-data resolution modes to monitors, giving more control over how missing data is handled in alerting.
• ›Enables sending web callouts directly from a trace or session view.
• ›Adds current user location context to the agent.

$ git clone --branch v3.182.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.182.0

$ git clone --branch v3.181.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.181.0