The daily firehose — everything the toolchain shipped today, already filtered.
// HOW THIS ISSUE IS MADE
We read every release from the 174 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
AutoGPT Platform v0.6.61 adds chat search, paginated session sidebar, and a redesigned profile popover.
└──▷ GET THIS VERSION
$ git clone --branch autogpt-platform-beta-v0.6.61 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:$ git checkout autogpt-platform-beta-v0.6.61
›Adds chat search modal behind the CHAT_SEARCH feature flag, enabling users to search across conversations.
›Paginates the session sidebar so conversation threads beyond 50 remain reachable.
›Redesigns the profile popover, including display of the user's avatar image.
›Adds a 'Response stopped' banner in Copilot to surface clearer feedback when a response is interrupted.
›Retires deprecated LLM models with family-aware migration, automatically moving users to supported model families.
langchain-fireworks 1.4.0 migrates to the fireworks-ai 1.x SDK and surfaces ContextOverflowError on prompt-too-long.
└──▷ GET THIS VERSION
$ git clone --branch langchain-fireworks==1.4.0 https://github.com/langchain-ai/langchain.git
# already have the repo? check out this version:$ git checkout langchain-fireworks==1.4.0
›Migrates the integration to the fireworks-ai 1.x SDK, enabling access to the updated Fireworks AI client surface.
›Raises ContextOverflowError when a prompt exceeds the model's context limit, giving callers a catchable, specific exception instead of a generic error.
└──▷ BREAKING ON UPGRADE
!The underlying client library is now fireworks-ai 1.x; any code that depended on internal APIs or behaviours of the pre-1.x SDK may break on upgrade.
OpenClaw v2026.5.19 adds typed plugin scaffolding, Android realtime voice, global skill installs, browser dialog handling, and sweeping QA-Lab expansions.
└──▷ GET THIS VERSION
$ git clone --branch v2026.5.19 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:$ git checkout v2026.5.19
└──▷ TRY IT
Bootstrap a new typed tool plugin project with manifest generation and context factory stubs — useful when building a custom tool you want to distribute.
›Adds defineToolPlugin plus openclaw plugins build, validate, and init for scaffolding typed simple tool plugins with generated manifest metadata, optional tool declarations, and context factories.
›Adds --global flag to openclaw skills install and openclaw skills update so shared managed skills can be targeted without per-user installs.
›Adds openclaw browser evaluate --timeout-ms to extend both the evaluate action and request timeout budgets for long-running in-page functions.
›Surfaces pending and recently handled browser modal dialogs in snapshots, returns blockedByDialog when an action opens a modal, and adds browser dialog --dialog-id to answer pending dialogs.
›Switches Android Talk Mode to realtime Gateway relay voice sessions with streaming mic input, realtime audio playback, tool-result bridging, and on-screen transcripts.
+18 moreshow less
›Adds openclaw qa suite --runtime-parity-tier to wire standard Codex-vs-Pi tier checks into release gates separately from optional/live-only/soak lanes.
›Exposes runtime tool fixture coverage through openclaw qa coverage --tools with optional suite-summary evaluation for parity gate artifacts.
›Adds /codex plugins list, enable, and disable for managing configured native Codex plugins from chat without editing config by hand.
›Adds OPENCLAW_IMAGE_APT_PACKAGES as the runtime-neutral image build arg for extra apt packages (with OPENCLAW_DOCKER_APT_PACKAGES kept as a legacy fallback).
›Adds OPENCLAW_IMAGE_PIP_PACKAGES for opt-in Python package installation in local Docker/Podman image builds.
›Supports HTTPS managed forward-proxy endpoints and scoped proxy.tls.caFile CA trust for proxy endpoint TLS.
›Exposes config lookup reload metadata so tools can distinguish restart-required, hot-reloadable, and no-op fields before applying config edits.
›Adds Telegram allowlisted native DM draft previews for transient tool progress while keeping final answers on the normal persistent delivery path.
›Adds admin HTTP RPC support for trusted clients to start and wait for web QR login flows.
›Adds a meme-maker skill covering curated template search, local SVG/PNG rendering, Imgflip hosted rendering, and Know Your Meme provenance links.
›Adds a Python debugging skill for pdb, breakpoint(), post-mortem inspection, and debugpy remote attach.
›Adds --output-schema flag to codex exec resume, enabling resumed automations to enforce structured JSON output while preserving session context.
›Enables remote executor registration via standard Codex auth, eliminating the need for a separate registry credential flow.
›Optimizes TUI startup by batching terminal capability probes, reducing time to first interactive frame.
›Supports preserving requested image fidelity — including original-resolution local images — across user inputs and image-producing tools in app-server turns.
OpenCode v1.15.6 adds a TUI diff viewer, shell mode in run, desktop tabs, and structured public error schemas in the v2 HTTP API.
└──▷ GET THIS VERSION
$ git clone --branch v1.15.6 https://github.com/sst/opencode.git
# already have the repo? check out this version:$ git checkout v1.15.6
›Adds a diff viewer in the TUI for reviewing file changes, with automatic focus on the first file and collapsed single-child directories in the file tree.
›Adds shell mode to the run prompt for running shell commands directly.
›Replaces subagent tabs with an on-demand picker in run for cleaner multi-agent workflows.
›Exposes structured public error schemas (including catalog and request errors) via the v2 HTTP API.
Pi v0.75.4 adds supply-chain hardening with shrinkwrap/lifecycle-script controls and interactive post-update changelogs.
└──▷ GET THIS VERSION
$ git clone --branch v0.75.4 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.75.4
└──▷ TRY IT
After upgrading, view the new version's changelog inline before resuming work.
$ pi update
›Supply-chain hardening: ships npm-shrinkwrap.json for transitive dependency locking, enforces dependency pinning and lifecycle-script allowlists, and disables lifecycle scripts during self-update and local release installs.
›Shows interactive update notes after pi update so users can review the installed version's changelog before continuing.
›Exports image resize utilities from the package root for SDK consumers.
Show symbol-style text icons next to completion menu items for quicker visual scanning of completion kinds.
json
{
"completion_menu_item_kind": "symbol"
}
›Adds subagent_model setting to specify which model is used when a subagent is spawned.
›Adds guardrail_identifier and guardrail_version settings for the AWS Bedrock provider, enabling use in environments where IAM policies require a guardrail on all model requests.
›Adds "completion_menu_item_kind": "symbol" config option to enable text/symbol icons for completion menu items (disabled by default).
›Adds vim.show_edit_predictions_in_normal_mode setting to control whether edit predictions are shown in normal mode.
›Adds support for opening a file at a line range in the file finder using path:start-end syntax (e.g. file.rs:10-20).
›Adds GGML_CUDA_PDL environment variable to enable or disable Programmatic Dependent Launch (PDL) on NVIDIA Hopper+ GPUs; set GGML_CUDA_PDL=0 to disable.
›Introduces PDL support for NVIDIA Hopper+ (SM90+) GPUs, overlapping CUDA kernel execution to improve inference throughput; enabled by default on qualifying devices, excluded on Ada and below.
llama.cpp b9253 introduces a unified llama executable with a serve subcommand for the server.
└──▷ GET THIS VERSION
$ git clone --branch b9253 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:$ git checkout b9253
›Introduces the llama unified executable, consolidating multiple tools under a single binary with subcommands including serve (for the server) and help.
v4.9 adds MTP speculative decoding, web search snippets, live token-speed display, drag-and-drop upload, and localhost-only CORS hardening.
└──▷ GET THIS VERSION
$ git clone --branch v4.9 https://github.com/oobabooga/textgen.git
# already have the repo? check out this version:$ git checkout v4.9
└──▷ TRY IT
Lock the API surface to localhost only (default after v4.9) or expose it to the network when needed.
$ # Default — localhost only, no extra flag needed
python server.py
# Opt into network exposure
python server.py --listen
# or for public API access
python server.py --public-api
›Adds draft-mtp as a new --spec-type option for MTP speculative decoding; auto-enabled when loading MTP GGUFs (e.g. Qwen 3.6 MoE MTP builds).
›Treats negative --ctx-size values as auto (0), allowing automatic context-size selection.
›Adds snippet support to the web_search tool so results include a short text excerpt, often eliminating the need for a follow-up fetch_webpage call.
›Shows live generation speed (tokens/s) and context size in the UI while generating.
›Adds drag-and-drop file upload support to the chat input.
+8 moreshow less
›Adds Linux aarch64 portable builds for DGX Spark support.
›Stores Electron app data in user_data/cache/electron instead of the OS default location.
›Auto-detects and auto-selects sibling mmproj-*.gguf files when loading a model; these appear in the mmproj dropdown and are hidden from the regular model dropdown.
›Drops link URLs from fetch_webpage output (links render as plain text instead of [text](url) markdown), significantly reducing tokens used per page.
›Adds 'Check for updates' button, a folder picker for the models directory, right-click copy context menu, and a spellcheck toggle in the Electron Session tab.
›Improves web search security by rejecting non-HTTP links.
›Hides reasoning and tools controls in chat mode (shown only in instruct / chat-instruct modes).
›One-click installer now tracks the latest release tag instead of bleeding-edge main.
└──▷ BREAKING ON UPGRADE
!Electron app data location changed from the OS default to user_data/cache/electron; existing cached data in the old location will not be picked up automatically.
vMLX 1.5.45 adds MCP config auto-discovery, chained tool calls across multi-round chats, and localized What's New notices.
└──▷ GET THIS VERSION
$ git clone --branch v1.5.45 https://github.com/jjang-ai/vmlx.git
# already have the repo? check out this version:$ git checkout v1.5.45
└──▷ TRY IT
Let vMLX auto-discover your MCP configuration without specifying a launch flag — just place your config in the working directory.
$ # Place your MCP config in the project root, then launch normally
# vMLX will auto-discover mcp.json or mcp.yaml without any extra flag
echo '{"mcpServers": {"my-server": {"command": "npx", "args": ["-y", "my-mcp-server"]}}}' > mcp.json
vmlx
›MCP config auto-discovery now loads mcp.json or mcp.yaml from the working directory or user config paths without requiring a launch flag, and MCP policy flows through sessions and gateway routes.
›Built-in coding tools and MCP tools can now chain across multi-round chats using structured tool_calls, with a text-chat watchdog that clears stuck 'Generating tool call' states.
›Adds support for Qwen3.6 native MTP, JANG/JANGTQ group-size detection, DSV4 cache policy, and TurboQuant/L2/paged cache telemetry with stricter release gates.
›The in-app What's New notice is now localized in English, Spanish, Japanese, Korean, and Chinese, covering MCP, MTP, cache, and updater safety notes.
THE AI TOOLCHAIN