The AI Toolchain — issue -002, August 17, 2026

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 v2.1.234 adds mid-turn command access, GitLab MR badges, auto-resume at usage limits, and a new CLAUDE_CODE_PROJECT_DIR_NAME env var.

$ git clone --branch v2.1.234 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.234

$ export CLAUDE_CODE_PROJECT_DIR_NAME=my-project
claude

$ export CLAUDE_CODE_GOAL_CHECKIN_MINUTES=0
claude

• ›Adds CLAUDE_CODE_PROJECT_DIR_NAME environment variable so hosts can set a short name for the per-project transcript directory when each session gets its own config directory.
• ›Adds CLAUDE_CODE_GOAL_CHECKIN_MINUTES environment variable to control how often /goal checks in on background tasks that have been waiting 30+ minutes (set to 0 to disable).
• ›Adds selection:clear keybinding action, bindable to a key to clear in-app text selections including in the agents view.
• ›Adds a GitLab merge request badge to the footer and statusline for repos with a GitLab remote and an authenticated glab CLI, showing MR !N with draft/pending/green states.
• ›Adds automatic session continuation when a claude.ai usage limit resets; toggle via /config with the 'Continue automatically at usage limit' setting.

Lightweight coding agent that runs in your terminal

Codex CLI gains endpoint protection detection in codex doctor, improved network diagnostics, and working-directory TUI commands.

$ git clone --branch rust-v0.148.0-alpha.21 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.148.0-alpha.21

const codex = new Codex({
  configOverrides: [
    '--config', 'sandbox.writable_roots."/var/myapp/data"=true'
  ]
});

• ›Improves codex doctor network diagnostics, including proxy environment details, macOS system proxy configuration detection, and rejection of proxy authentication challenges (HTTP 407).
• ›Adds CodexOptions.configOverrides to the TypeScript SDK for passing ordered --config key=value arguments to Codex CLI unchanged, enabling TOML permission maps with literal path keys.
• ›Adds working-directory commands to the TUI.
• ›Supports per-environment shell variable policies and environment-specific command policies.
• ›Adds experimental app-server project APIs.

OpenHands: AI-Driven Development

OpenHands v1.14.0 adds Git Sync automation, LLM pre-flight validation, structured Canvas error outcomes, and sets Kimi K3 as the free default Canvas model.

$ git clone --branch v1.14.0 https://github.com/All-Hands-AI/OpenHands.git
# already have the repo? check out this version:
$ git checkout v1.14.0

• ›Adds a Git Sync page under Automations for managing repository synchronization workflows.
• ›Adds LLM pre-flight validation when saving LLM profiles, blocking misconfigured settings before they are persisted.
• ›Canvas now presents structured error outcomes, surfacing failure details in a readable format rather than raw output.
• ›Sets Kimi K3 as the default Canvas model and tags it as free.

$ git clone --branch v1.14.0 https://github.com/OpenHands/OpenHands.git
# already have the repo? check out this version:
$ git checkout v1.14.0

Qwen Code v0.21.13 adds conversation forking, file drag-and-drop in Web Shell, --since flag for incremental PR review, and a new External Context Provider Extension Profile.

$ git clone --branch v0.21.13 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.21.13

$ qwen review fetch-pr --since <ledger-marker> <pr-number>

$ qwen findings --to-anchors <findings-file>

• ›Adds --since flag to qwen review fetch-pr for validating incremental review anchors and scoping diffs based on local cache or ledger markers.
• ›Adds --to-anchors option to the findings pipeline to normalize inputs for final gates and accept bracketed source tags.
• ›Introduces the External Context Provider Extension Profile v1 to enable provider-owned retrieval integrations via Qwen Extension and MCP boundaries.
• ›The /review skill now uses dedicated platform subcommands meta and fetch-diff instead of executing raw gh commands.
• ›Stopped takeover PRs now receive an autofix/needs-human label and appear in a new Takeover pool table on the dashboard.

The agent that grows with you

Hermes v2026.8.16.2 adds Bot Mode plugin, Cua Driver 0.20, remote gateway headers, /rollback --all, and raised Codex OAuth context limits.

$ git clone --branch v2026.8.16.2 https://github.com/NousResearch/hermes-agent.git
# already have the repo? check out this version:
$ git checkout v2026.8.16.2

• ›Bundles Bot Mode (hermes-bots) as a built-in, default-on desktop plugin implementing the core teammate protocol for Bot Chat sessions.
• ›Supports Cua Driver 0.20 runtime contracts for computer-use, including auto-repair of installed drivers that fail the runtime contract at update or runtime.
• ›Adds remote gateway headers support in the desktop app, carried through the connections registry, test probes, and Settings UI.
• ›Adds status bar reconnect for offline gateways and self-healing for dropped SSH/HTTP registered remote connections in the desktop.
• ›Adds connection-aware plugin routing in the desktop, scoping plugin socket connections and session/pin lists per connection.

ONNX Runtime ships its first CUDA Plugin EP release, delivering CUDA execution as a separately packaged, independently installable provider.

$ git clone --branch plugin-ep-cuda/v0.1.0 https://github.com/microsoft/onnxruntime.git
# already have the repo? check out this version:
$ git checkout plugin-ep-cuda/v0.1.0

• ›Introduces the CUDA Plugin EP as a separately packaged execution provider, now the default CUDA provider implementation for ONNX Runtime, distributed as the onnxruntime-ep-cuda Python wheel and NuGet packages.
• ›Adds provider options for user compute streams, copy behavior, EP-level unified streams, and external allocators, with compatibility back to ONNX Runtime 1.24.4 via version-gated callbacks.
• ›Adds plugin profiling APIs and memory statistics in profiling output.
• ›Adds CUDA Graph capture and replay, including user compute stream support, synchronization controls, and a kernel sync stream for scratch allocation.
• ›Adds arena allocation, resource accounting, available-resource reporting, and IOBinding synchronization.

Arize Phoenix v20.3.0 adds an expression filter DSL for querying traces.

$ git clone --branch arize-phoenix-v20.3.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v20.3.0

• ›Adds an expression filter DSL for traces, enabling structured query filtering over trace data.

Langfuse v4.12.0 adds a v4 migration UI, filter sidebar search, PostHog export failure notifications, and off-thread trace graph rendering.

$ git clone --branch v4.12.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v4.12.0

• ›Adds a v4 migration UI toggle in the settings page and a migration link in personal settings menus, letting teams self-serve the v4 data-layer transition.
• ›Adds a PostHog export failure project notification event so teams are alerted when the PostHog analytics export breaks.
• ›Surfaces persisted PostHog export faults directly in the settings page, making misconfiguration visible without log-diving.
• ›Adds filter sidebar search by facet name, letting users quickly locate specific filters across large facet lists.
• ›Moves large trace graph layout off the main thread, unblocking interactive exploration of traces with many spans.