The AI Toolchain — issue -104, May 7, 2026

AutoGPT Platform v0.6.59 adds tier-based file storage, admin CSV exports, yearly Stripe billing, and a new get_platform_info tool for AutoPilot.

$ git clone --branch autogpt-platform-beta-v0.6.59 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:
$ git checkout autogpt-platform-beta-v0.6.59

• ›Adds get_platform_info tool for tier-aware AutoPilot, enabling agents to query platform tier information at runtime.
• ›Adds dynamic max_budget_usd for the SDK with parity budget hint for baseline runs.
• ›Introduces admin CSV exports for credit transactions and copilot weekly usage.
• ›Adds tier-based workspace file storage limits.
• ›Adds yearly Stripe billing with pricing and frontend hookup.

The agent that grows with you

Hermes v2026.5.7 ships durable multi-agent Kanban, /goal persistence, video analysis, 8 P0 security closures, and Google Chat as platform #20.

$ git clone --branch v2026.5.7 https://github.com/NousResearch/hermes-agent.git
# already have the repo? check out this version:
$ git checkout v2026.5.7

$ /goal Enumerate all exposed S3 buckets in the org and produce a findings report

$ hermes curator archive && hermes curator prune

• ›Adds durable multi-agent Kanban board with heartbeats, zombie detection, per-task retries, hallucination gate, and multi-project support across profiles.
• ›New /goal command locks the agent onto a target across turns (Ralph loop as a first-class primitive).
• ›New video_analyze tool enables native video understanding on Gemini and compatible multimodal models.
• ›Adds xAI Custom Voices as a TTS provider with voice cloning support.
• ›Adds Google Chat as the 20th supported messaging platform; introduces generic platform-plugin hooks so third-party adapters drop in without touching core.

Build resilient agents.

LangGraph CLI gains Studio deploy support for pushing graphs directly to LangGraph Studio.

$ git clone --branch cli==0.4.25 https://github.com/langchain-ai/langgraph.git
# already have the repo? check out this version:
$ git checkout cli==0.4.25

• ›Adds studio deploy command to the LangGraph CLI, enabling direct deployment to LangGraph Studio.

OpenAI Agents SDK v0.16.0 adds MCP server-prefixed tool names, per-run tool concurrency config, and an unlimited-turns mode.

$ git clone --branch v0.16.0 https://github.com/openai/openai-agents-python.git
# already have the repo? check out this version:
$ git checkout v0.16.0

agent = Agent(
    name="Assistant",
    mcp_servers=[my_mcp_server],
    mcp_config={"include_server_in_tool_names": True},
)

$ OPENAI_DEFAULT_MODEL=gpt-4.1 python my_agent.py

• ›Adds include_server_in_tool_names to MCPConfig (set True) to prefix each MCP tool name with its server name, preventing collisions when multiple MCP servers expose identically named tools.
• ›Adds ToolExecutionConfig(max_function_tool_concurrency=...) on RunConfig to cap SDK-side local function tool execution concurrency independently of the provider-side ModelSettings.parallel_tool_calls setting.
• ›Adds max_turns=None to the run API to disable the turn limit entirely, while preserving the existing default of DEFAULT_MAX_TURNS (10) when max_turns is omitted.
• ›Adds OPENAI_DEFAULT_MODEL environment variable as a global override for the SDK default model, allowing the previous gpt-4.1 behavior to be restored without per-agent code changes.

PydanticAI v1.91.0 adds gpt-image-2 options for OpenAI and support for deepseek-v4-flash and deepseek-v4-pro models.

$ git clone --branch v1.91.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.91.0

• ›Supports gpt-image-2 model options via the OpenAI provider.
• ›Adds deepseek-v4-flash and deepseek-v4-pro to the DeepSeek provider.

SRE Agent - CNCF Sandbox Project

HolmesGPT 0.28.0 adds DNS configuration support and authenticated private-channel realtime subscriptions.

$ git clone --branch 0.28.0 https://github.com/HolmesGPT/holmesgpt.git
# already have the repo? check out this version:
$ git checkout 0.28.0

• ›Adds the ability to configure dnsPolicy and dnsConfig for HolmesGPT deployments.
• ›Enables private channel support with authentication for realtime subscriptions.

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 v2.1.133 adds worktree base-ref control, custom sandbox binary paths, admin policy merge, and effort-level hook inputs.

$ git clone --branch v2.1.133 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.133

$ #!/bin/bash
# In a Claude hook script
echo "Current effort: $CLAUDE_EFFORT"
if [ "$CLAUDE_EFFORT" = "low" ]; then
  exit 0  # skip expensive checks
fi

• ›Adds worktree.baseRef setting (fresh | head) to control whether new worktrees branch from origin/<default> or local HEAD.
• ›Adds sandbox.bwrapPath and sandbox.socatPath managed settings (Linux/WSL) to specify custom bubblewrap and socat binary locations.
• ›Adds parentSettingsBehavior admin-tier key ('first-wins' | 'merge') to opt SDK managedSettings into the policy merge chain.
• ›Hooks now receive the active effort level via the effort.level JSON input field and the $CLAUDE_EFFORT environment variable, readable by Bash tool commands.

Lightweight coding agent that runs in your terminal

Codex CLI v0.129.0 adds Vim composer mode, /hooks browser, /ide context injection, workspace-aware /diff, and a /keymap debug inspector.

$ git clone --branch rust-v0.129.0 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.129.0

$ /vim

$ /hooks

$ /keymap debug

• ›Adds modal Vim editing in the TUI composer via /vim command, configurable default mode, and Vim-specific keymap contexts.
• ›Adds /hooks browser for browsing and toggling lifecycle hooks, with support for pre/post-compaction hooks and PreToolUse context.
• ›Adds /ide context injection so TUI workflows can pull in IDE-side context directly.
• ›Adds workspace-aware /diff command for comparing workspace state within TUI sessions.
• ›Adds /keymap debug inspector to inspect and diagnose terminal key events.

The open source coding agent.

OpenCode v1.14.41 lets session warps carry uncommitted changes and adds ACP session state restore.

$ git clone --branch v1.14.41 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.14.41

• ›Warping a session to another workspace now carries over uncommitted file changes.

$ git clone --branch v1.14.40 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.14.40

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

Pi v0.73.1 adds self-update npm scope migration support, interactive OAuth login selection, and JSONC-style models.json parsing.

$ git clone --branch v0.73.1 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.73.1

$ pi update --self

• ›Adds npm scope migration support to pi update --self: uninstalls the old @mariozechner/pi-coding-agent package and installs the renamed @earendil-works/pi-coding-agent package automatically.
• ›Enables interactive OAuth login selection so providers can present multiple login choices in /login for provider-specific authentication flows.
• ›Supports JSONC-style models.json parsing — comments and trailing commas are now allowed, making custom provider and model configuration easier to maintain.

Qwen Code v0.15.8 adds a light theme toggle to HTML exports and replaces the agent display with an always-on LiveAgentPanel.

$ git clone --branch v0.15.8 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.15.8

• ›Adds light theme and theme toggle to /export HTML output via the web-templates feature.
• ›Replaces the inline AgentExecutionDisplay with an always-on LiveAgentPanel in the CLI for persistent agent status visibility.

$ git clone --branch v0.15.7 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.15.7

$ qwen review

$ /tasks

Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Ollama v0.23.2 caches /api/show for ~6.7x faster model-load latency and refines launch integration management.

$ git clone --branch v0.23.2 https://github.com/ollama/ollama.git
# already have the repo? check out this version:
$ git checkout v0.23.2

$ ollama launch claude-desktop --restore

• ›Caches /api/show API responses, cutting median latency by ~6.7x and accelerating integrations such as VS Code.
• ›Adds ollama launch claude-desktop --restore command to return Claude Desktop to its normal state after integration removal.
• ›Improves backup workflow when managing ollama launch integrations.
• ›Delivers a cleaner image-generation layout in the MLX runner.

Adds OpenCL operation-filter regex for targeted GPU kernel debugging in llama.cpp b9063.

$ git clone --branch b9063 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b9063

• ›Adds an opfilter regex to the OpenCL backend to selectively enable or disable specific GPU operations during debugging.

$ git clone --branch b9060 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b9060

$ git clone --branch b9055 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b9055

oobabooga textgen v4.8 adds a --no-electron flag, list-format API content support, and a redesigned chat composer.

$ git clone --branch v4.8 https://github.com/oobabooga/textgen.git
# already have the repo? check out this version:
$ git checkout v4.8

$ textgen --no-electron

• ›Adds --no-electron flag to skip the Electron desktop window and use the web UI in the browser instead.
• ›API now supports list-format content in tool and assistant messages.
• ›Redesigned chat composer with a taller input area and pinned paperclip and message-action buttons.
• ›Electron window now persists bounds and maximize state across launches.
• ›Smooth scroll animation when sending a new message in the chat UI.

Composio [email protected] adds Tool Router v3.1 with preload and session.update(), plus REVOKED connected account status and sandbox compute tiers.

$ git clone --branch [email protected] https://github.com/ComposioHQ/composio.git
# already have the repo? check out this version:
$ git checkout [email protected]

• ›Adds Tool Router v3.1 support including preload, SDK-local custom tool preload, attach/use flow refinements, and session.update().
• ›Adds REVOKED status for connected accounts, along with connected account string/list coercion and the allow_multiple guard.
• ›Adds workbench sandbox compute tier support.

$ git clone --branch @composio/[email protected] https://github.com/ComposioHQ/composio.git
# already have the repo? check out this version:
$ git checkout @composio/[email protected]

$ composio local-tools doctor

$ composio local-tools list

$ git clone --branch @composio/[email protected] https://github.com/ComposioHQ/composio.git
# already have the repo? check out this version:
$ git checkout @composio/[email protected]

const session = await composio.use(existingSessionId);
await session.update({ connectedAccounts: { gmail: 'ca_new_xxx' } });

const session = await composio.use(existingSessionId, {
  customTools: myCustomTools,
  customToolkits: myCustomToolkits
});
const tools = await session.tools();

import { SessionPreset } from '@composio/core';

const session = await composio.create({
  preset: SessionPreset.DIRECT_TOOLS,
  preload: { tools: 'all' }
});
const tools = await session.tools();