The AI Toolchain — issue -036, July 14, 2026

Agno v2.7.3 adds ValkeyDb storage and vector store, RedmineTools, TokenLab provider, and AG-UI human-in-the-loop support

$ git clone --branch v2.7.3 https://github.com/agno-agi/agno.git
# already have the repo? check out this version:
$ git checkout v2.7.3

• ›Adds ValkeyDb as a fast in-memory database backend for agents, teams, and workflows.
• ›Adds ValkeyDB vector store with both vector and keyword search capabilities.
• ›Adds RedmineTools to manage issues, comments, and time logs against a Redmine project management instance.
• ›Adds TokenLab as a new OpenAI-compatible model provider.
• ›Extends AG-UI with human-in-the-loop confirmation, input, and feedback flows.

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 v2.1.210 adds live tool-call timers, permission-rule warnings, screen reader announcements, and smarter agent dashboard cues.

$ git clone --branch v2.1.210 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.210

• ›Adds a live elapsed-time counter to the collapsed tool summary line so long-running tool calls visibly tick instead of appearing stuck.
• ›Adds a startup warning when Write(path), NotebookEdit(path), or Glob(path) are used as permission rules, directing users to Edit(path) or Read(path) instead.
• ›Screen reader mode now announces permission mode changes aloud when cycling modes with Shift+Tab.
• ›Memory writes that leave a MEMORY.md index over its read limit now produce an explicit error instead of silent truncation.
• ›The agents footer hint now shows how many background agents are waiting on input, with color emphasis when the count changes.

$ git clone --branch v2.1.208 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.208

$ CLAUDE_AX_SCREEN_READER=1 claude

$ CLAUDE_CODE_PROCESS_WRAPPER=/usr/local/bin/corp-launcher claude

# In ~/.claude/settings.json
{ "vimInsertModeRemaps": { "jj": "Escape" } }

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

Goose v1.43.0 adds per-message cost/token stats, new AI providers, TLS for ACP serve, and a session --edit flag for conversation forking.

$ git clone --branch v1.43.0 https://github.com/block/goose.git
# already have the repo? check out this version:
$ git checkout v1.43.0

$ goose session --edit <session-id>

• ›Adds --edit flag to session commands to edit conversation history before forking a new session.
• ›New per-message usage stats UI showing tokens, cost, time-to-first-token, and tokens/sec.
• ›Per-message usage and cost tracking with derived session totals.
• ›Groups chat sessions by project in the navigation panel.
• ›Shows project names in the directory switcher dropdown.

The open source coding agent.

OpenCode Desktop v1.18.0 adds a layout switcher and completes the v2 migration with first-launch onboarding.

$ git clone --branch v1.18.0 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.18.0

• ›Adds a setting to switch between the new (v2) and legacy Desktop layouts during the transition period.
• ›Completes the Desktop v2 migration, including upgrade handling and first-launch onboarding flow.

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

Pi v0.80.7 adds cache-friendly dynamic tool loading, Fable 5 xhigh/max thinking, Ctrl+X message copy, and toolChoice for OpenAI/Codex Responses.

$ git clone --branch v0.80.7 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.80.7

$ # In the Pi TUI, navigate to /tree, select the desired message, then press Ctrl+X

• ›Adds cache-friendly dynamic tool loading: extensions can inject tools during execution while Anthropic and OpenAI Responses models preserve prompt-cache prefixes.
• ›Adds Ctrl+X shortcut to copy the last assistant message or the selected message in /tree, enabling direct copying of branched and historical messages.
• ›Adds native xhigh and max thinking levels for Claude Fable 5 across all generated provider catalogs.
• ›Adds toolChoice support (including required and named tool selection) for OpenAI and Codex Responses providers.

Qwen Code SDK v0.1.8 adds session resumption, multi-modal input, concurrent batch execution, MCP progress updates, and a Coding Plan auth mode.

$ git clone --branch sdk-typescript-v0.1.8 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout sdk-typescript-v0.1.8

$ qwen --session-id <your-session-id> 'continue the refactor we started'

{
  "extra_body": { "enable_thinking": true }
}

env:
  MY_API_KEY: "sk-..."
  HTTP_PROXY: "http://proxy.corp:8080"

• ›Adds --session-id CLI flag and resume/continue SDK options to resume or continue existing sessions by ID.
• ›Adds settings.env field for environment variable configuration directly in settings.
• ›Adds contextWindowSize configuration key to control the model context window.
• ›Adds extra_body support for OpenAI-compatible providers to pass provider-specific request fields.
• ›Adds export command for session history in markdown and HTML formats.

$ git clone --branch v0.19.10 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.19.10

LocalAI 4.7.0 adds voice cloning profiles, local video/avatar generation, diarized transcription, streaming TTS, and auto full-context resolution.

$ git clone --branch v4.7.0 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:
$ git checkout v4.7.0

context_size: -1

options:
  - devices:CUDA1,CUDA2

$ curl -X POST http://localhost:8080/v1/audio/speech \
  -H 'Content-Type: application/json' \
  -d '{"model": "qwen3-tts-cpp", "input": "Hello, world!", "voice": "localai://voice-profiles/my-profile"}'

• ›Adds GET /api/voice-profiles, POST /api/voice-profiles, DELETE /api/voice-profiles/:id, and GET /api/voice-profiles/:id/audio REST endpoints for admin-managed voice cloning profiles; profiles are referenceable via stable localai://voice-profiles/<id> URIs from /v1/audio/speech or the TTS UI.
• ›Adds MCP admin tools list_voice_profiles, create_voice_profile, and delete_voice_profile for voice profile management.
• ›Adds typed tts.voice_cloning boolean config key to opt custom model names into or out of Voice Library compatibility; setting it to false rejects saved profile references with HTTP 400.
• ›Adds context_size: -1 sentinel (any negative value, or LOCALAI_CONTEXT_SIZE=-1 / --context-size -1 globally) to resolve a model's context window to its full trained n_ctx_train from GGUF metadata at load time, with a VRAM-fit warning and a safe clamp so no backend receives a negative value.
• ›Adds devices: option to the llama.cpp options: array (e.g., options: [devices:CUDA1,CUDA2]) to restrict GPU offload to specific named devices from llama-server --list-devices.

llama.cpp b10010 adds --cors-* CLI options to the server for fine-grained CORS control.

$ git clone --branch b10010 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b10010

• ›Adds --cors-* options to the server command, including a special localhost value, enabling fine-grained cross-origin resource sharing control.

$ git clone --branch b10003 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b10003

$ echo 'Analyze this payload' | llama-tokenize --stdin -hf <org/model> --show-count

$ llama-tokenize -m /models/llama.gguf --offline --no-bos --ids -f input.txt

$ git clone --branch b9994 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b9994

Phoenix v18 shifts session time-range filters to interval-overlap semantics.

$ git clone --branch arize-phoenix-v18.0.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v18.0.0

• ›Session time-range filters now use interval-overlap semantics, returning sessions where any part of the session overlaps the queried range rather than requiring full containment.

$ git clone --branch arize-phoenix-v17.30.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-v17.30.0

Langfuse v3.213.0 enables monitors for self-hosted deployments and adds a configurable REDIS_SOCKET_TIMEOUT_MS environment variable.

$ git clone --branch v3.213.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:
$ git checkout v3.213.0

$ REDIS_SOCKET_TIMEOUT_MS=10000 docker compose up

• ›Adds REDIS_SOCKET_TIMEOUT_MS environment variable to make Redis socket timeout configurable for self-hosted deployments.
• ›Enables monitors for self-hosted Langfuse deployments via event writes, bringing parity with cloud-hosted monitoring.
• ›Adds contract-aware autocomplete to the code evaluator editor, surfacing available variables and types inline while writing eval logic.
• ›Adds a copy-session-id button to the session detail header for quick clipboard access to session identifiers.
• ›Writes a run-completion manifest per export run in blob storage, enabling downstream consumers to detect when an export run is complete.

LanceDB python-v0.35.0-beta.2 adds a blob v2 fetch API, WatsonxReranker support, and FTS query tokenization.

$ git clone --branch python-v0.35.0-beta.2 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout python-v0.35.0-beta.2

• ›Adds WatsonxReranker component for reranking search results using IBM Watsonx.
• ›Adds FTS query tokenization support for table full-text-search queries.
• ›Introduces a blob v2 fetch API for retrieving binary large object data.

$ git clone --branch v0.32.0-beta.2 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout v0.32.0-beta.2

Milvus v2.6.20 improves query scheduling, batching, index loading, filter performance, and streaming rebalancing with observability upgrades.

$ git clone --branch v2.6.20 https://github.com/milvus-io/milvus.git
# already have the repo? check out this version:
$ git checkout v2.6.20

• ›Adds named C++ thread-pool activity metrics with Grafana monitoring support.
• ›Enables the channel-level score balancer by default and introduces a safer default threshold for channel-exclusive mode, improving streaming rebalancing.
• ›Improves QueryCoord scheduling by decoupling task dispatch from distribution polling to allow independent scheduling intervals.
• ›Improves QueryNode query batching by increasing the default NQ grouping limits for larger merged query batches.
• ›Improves filter execution performance by skipping null-bitmap processing for all-valid results.