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.
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 v2.1.191 adds /rewind to restore conversations before /clear, cuts streaming CPU usage ~37%, and hardens MCP reliability.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.191 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.191
└──▷ TRY IT
Accidentally cleared your conversation mid-session? Use /rewind to jump back to the state before /clear and continue without losing context.
$ /rewind
›Adds /rewind command to resume a conversation from the state it was in before /clear was run.
›Sandbox network permission dialog now remembers allowed hosts for the entire session, eliminating repeated prompts.
›MCP capability discovery (tools/list, prompts/list, resources/list) now retries transient network errors with backoff, improving server reliability.
›MCP OAuth in headless environments skips the browser popup and goes directly to the paste-the-URL prompt.
›MCP HTTP 404 errors now surface the URL and point to the MCP config file for faster debugging.
GitHub Copilot CLI v1.0.65 adds a copilot skill subcommand, CI check status bar, and shell-history search across modes.
└──▷ GET THIS VERSION
$ git clone --branch v1.0.65 https://github.com/github/copilot-cli.git
# already have the repo? check out this version:$ git checkout v1.0.65
└──▷ TRY IT
List all currently active skills to audit what capabilities are loaded.
$ copilot skill list
Enable the CI status bar item to monitor branch check runs without leaving the CLI.
📍In the CLI, open /settings and enable the CI check status bar item to display passing/running/failing state for the current branch.
›Adds copilot skill subcommand (and /skill alias for /skills) to list, add, and remove skills from a file, URL, or directory.
›Adds an opt-in status bar item showing CI check status (passing/running/failing) for the current branch.
›Up/down history and Ctrl+R reverse search now include past shell commands while in normal mode, so shell commands can be recalled without first typing ! to enter shell mode.
›/cd now persists the working directory so resuming a session returns to it, and discovers custom agents in the new directory.
›Saves custom status line commands in /settings.
+5 moreshow less
›Open canvases resume automatically after restarting the CLI.
›Includes userPromptSubmitted hook additionalContext in the model-facing prompt.
›Parses /every schedules on the session's main model.
›Silent MCP OAuth refresh reuses the granted scope so reconnects stay signed in.
›Keeps custom-agent subagent model selections when using BYOK providers.
OpenHands 1.39.0 adds ACP model switching, MCP server config for ACP agents, Slack attachment context, multi-model LLM discovery, copyable invite links, and a sub-agent task visualizer.
└──▷ GET THIS VERSION
$ git clone --branch cloud-1.39.0 https://github.com/All-Hands-AI/OpenHands.git
# already have the repo? check out this version:$ git checkout cloud-1.39.0
└──▷ TRY IT
Pin a custom resolver label per deployment so automation events are routed to the correct resolver macro.
$ OH_RESOLVER_LABEL=my-org-resolver
Copy and share an org invite link directly from the console so new members can join without waiting for an email.
📍In the console, go to Settings › Organization › Members and click "Copy invite link" to share a URL that auto-accepts on click.
›Adds switch_acp_model proxy endpoint enabling model switching for cloud ACP conversations.
›Adds model dropdown and credential fields to the agent settings page for ACP agents.
›Supports MCP server configuration for ACP agents.
›Enables multi-model LLM discovery and BYOK gating for OpenHands Enterprise (OHE) installs, including proxy discovery and catalogue union.
›Adds hide_personal_workspaces flag to restrict OHE installs to org-only workspaces.
+16 moreshow less
›Adds explicit OH_DEPLOYMENT_MODE flag to control deployment mode and gate cloud-account CTAs to enterprise cloud.
›Adds OH_RESOLVER_LABEL env var to configure a per-deployment resolver macro.
›Exposes copyable invite links so org invitations work without email delivery.
›Auto-accepts invitation links on click, removing the confirm/cancel modal.
›Accepts pending invitations by verified-email match at sign-in.
›Adds validation for secret names to ensure valid environment variable format.
›Surfaces Slack attachments as context in agent conversations.
›Adds a sub-agent task (TaskToolSet) visualizer to the chat UI.
›Adds DynamicRemoteSandboxSpecService backed by runtime-API warm configs for dynamic sandbox provisioning.
›Parallelizes GitHub GraphQL queries in get_suggested_tasks for faster task suggestions.
›Surfaces a clear error in the UI when a custom sandbox image's agent-server SDK version mismatches.
›Persists and injects per-user OAuth tokens in Jira Data Center resolver conversations.
›Exposes the Automations menu item to all users.
›Routes unclaimed automation events to the default org on single-org installs.
›Moves existing users into the default org when first auto-added, and makes the first signed-in user the default org owner (tracked via is_default flag).
›Encodes agent harness in the conversation chip icon and surfaces the model name in chip text.
Build a dev container using the classic Docker builder when working on Apple Container or another engine without integrated BuildKit.
json
{
"dev_container_use_buildkit": false
}
Select the contents of the enclosing brackets at the cursor, then expand outward through nested pairs by repeating the action.
📍editor: select inside delimiters
›Adds the agent.terminal_init_command setting to automatically run a command when opening a new agent terminal thread in the agent panel.
›Adds a dev_container_use_buildkit setting to build dev containers with the classic Docker builder for engines without an integrated BuildKit (e.g. Apple Container).
›Adds editor: select inside delimiters and editor: select around delimiters actions for selecting content within the nearest enclosing brackets, braces, parentheses, or quotes; repeating expands the selection to the next enclosing pair.
›Adds a workspace: reset pane sizes command that equalizes the sizes of all panes in the center group.
›Adds a zed: rebuild dev extension action to rebuild an installed dev extension from the command palette.
+9 moreshow less
›Adds the ability to grant agent terminal commands network access to specific hosts (rather than all-or-nothing), enforced by an allowlisting proxy.
›Enables creating a new worktree directly from the sidebar's 'new thread' icon button.
›Adds Anthropic-compatible provider support in settings.
›Adds a 'View File' option to the Git Panel context menu to open a changed file in the editor without a diff view.
›Adds a Stop button to REPL notebook cells to interrupt the running kernel while a cell is executing.
›Prettier now reads the project's .editorconfig files when formatting.
›Improves the archive 'Search all threads' view to support filtering by project name, not just thread title.
›Adds the file icon to breadcrumbs when the tab bar is hidden, if file icons are enabled for tabs.
›Pressing up in an empty agent message editor now recalls the last queued message for editing.
$ git clone --branch v0.17.7 https://github.com/openai/openai-agents-python.git
# already have the repo? check out this version:$ git checkout v0.17.7
›Exposes a configurable max_size limit for WebSocket connections, allowing callers to raise or lower the message-size cap.
›Adds buffered Chat Completions tool-call streaming, delivering complete tool-call payloads as a single event rather than fragmenting them across stream chunks.
OpenClaw v2026.6.10 adds automatic fast mode, a session-transcript SDK, expanded GLM-5.2 reasoning tiers, and new OpenCode Go catalog models.
└──▷ GET THIS VERSION
$ git clone --branch v2026.6.10 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:$ git checkout v2026.6.10
└──▷ TRY IT
Enable automatic fast mode so short exchanges get fast responses while complex or fallback work transparently switches back to normal mode — useful for mixed interactive/agentic workflows.
$ openclaw /fast auto
›Adds /fast auto command so short conversational calls start in fast mode and longer or fallback work automatically returns to normal mode with effective state still visible in status.
›Displays the effective automatic fast-mode state in status (instead of a simple on/off) so operators can see what a session is actually using, including in connected-agent sessions.
›Adds a durable session-transcript SDK contract so plugins can read, append, publish, and lock the intended transcript without relying on legacy file paths as identity.
›Expands zai/glm-5.2 thinking choices beyond binary on/off, enabling high or max reasoning effort requests to Z.AI.
›Adds GLM-5.2 and Kimi K2.7 Code to the OpenCode Go model catalog so users can select them directly from OpenClaw.
+2 moreshow less
›Enables ClawHub discovery for the StepFun provider plugin so operators can install it through either ClawHub or npm.
›Keeps openclaw doctor core diagnostics in their normal order before extension checks, making diagnostic output easier to follow.
Hexagon NPU backend gains 32×32 tiled weight repack, kernel-param caching, and GGML_HEXAGON_MM_SELECT for matmul dispatch control.
└──▷ GET THIS VERSION
$ git clone --branch b9784 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:$ git checkout b9784
›Adds GGML_HEXAGON_MM_SELECT environment variable to control matmul kernel selection on Hexagon NPU, selectable from the host.
›Introduces simple graph caching on the Hexagon backend to avoid recomputing kernel-params on repeated graph executions.
›Implements 32×32 tiled weight repack layout for HMX matmul kernels, aligning tiles with DMA for higher throughput.
›Adds non-tiled MUL_MAT fallback path on Hexagon, falling back to HVX flat kernels when HMX tiling is not viable.
›Drops support for Hexagon arch < v73; HMX is now required for the backend — non-v73+ devices will no longer work.
└──▷ BREAKING ON UPGRADE
!Support for Hexagon architecture versions below v73 has been removed; HMX is now required, so devices running arch < v73 will no longer function with the Hexagon backend.
Phoenix v17.11.0 adds agent-readable eval failure reports, trace annotation summaries, and Swagger UI exposure of agent endpoints in dev mode.
└──▷ GET THIS VERSION
$ git clone --branch arize-phoenix-v17.11.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:$ git checkout arize-phoenix-v17.11.0
└──▷ TRY IT
Explore agent endpoints interactively via Swagger UI during local development without needing external API documentation.
$ python -m phoenix.server.main --dev
›Exposes agent endpoints in Swagger UI when running the server with --dev, making it easier to explore and test agent APIs interactively.
›Adds agent-readable failure reports for the PXI eval harness, surfacing structured evaluation failures directly to agents.
›Shows trace annotation summaries in the project stats panel, giving a at-a-glance view of annotation coverage across traces.
›Enables submitting a highlighted command on Enter in the agent skill prompt UI.
THE AI TOOLCHAIN