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 v2.1.108 adds 1-hour prompt cache TTL opt-in, session recap, and slash-command discovery via Skill tool.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.108 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.108
└──▷ TRY IT
Opt into the 1-hour prompt cache TTL on Bedrock to cut repeated-context costs across long sessions.
$ ENABLE_PROMPT_CACHING_1H=1 claude
Re-orient yourself after stepping away from a long session without scrolling through the full transcript.
$ /recap
Resume work in the current repo — the picker now defaults to sessions from this directory; switch to all projects if needed.
$ /resume # Ctrl+A to expand to all projects
›Adds ENABLE_PROMPT_CACHING_1H env var to opt into 1-hour prompt cache TTL across API key, Bedrock, Vertex, and Foundry backends; FORCE_PROMPT_CACHING_5M forces the 5-minute TTL.
›Adds /recap command (and CLAUDE_CODE_ENABLE_AWAY_SUMMARY env var) to generate a context summary when returning to a session; also configurable via /config.
›Enables the model to discover and invoke built-in slash commands (/init, /review, /security-review) via the Skill tool.
›Adds /undo as an alias for /rewind.
›Adds a warning at startup when prompt caching is disabled via DISABLE_PROMPT_CACHING* environment variables.
+5 moreshow less
›Adds a 'verbose' indicator when viewing the detailed transcript via Ctrl+O.
›Improves /resume picker to default to sessions from the current directory; Ctrl+A reveals all projects.
›Improves /model to warn before switching models mid-conversation to flag the uncached full-history re-read cost.
›Improves error messages: rate limits distinguished from plan usage limits; 5xx/529 errors link to status.claude.com; unknown slash commands suggest the closest match.
›Reduces memory footprint for file reads, edits, and syntax highlighting by loading language grammars on demand.
└──▷ BREAKING ON UPGRADE
!ENABLE_PROMPT_CACHING_1H_BEDROCK is deprecated (still honored) in favor of the unified ENABLE_PROMPT_CACHING_1H env var.
›Plugin hooks now receive PLUGIN_ROOT, COPILOT_PLUGIN_ROOT, and CLAUDE_PLUGIN_ROOT environment variables pointing to the plugin's installation directory.
›ACP server now binds to localhost only, preventing unintended network exposure.
›Renames 'steering' to 'remote control' in --remote flag and /remote command help text.
›Remote tab now shows Copilot coding agent tasks and supports remote control without requiring a pull request.
›Enterprise login now accepts hostnames without a URL scheme (e.g. github.example.com).
+4 moreshow less
›Instruction files with specific applyTo patterns are consolidated into a table instead of inlining full content, reducing context window usage.
›ctrl+o now expands all timeline entries, matching the behavior of ctrl+e.
›Adds deprecation warning when installing plugins from repos, URLs, or local paths.
›Session scope selector in sync prompt is now keyboard-navigable with left/right arrow keys.
Pi v0.67.2 adds multi-flag system prompt appending, Kitty super-key bindings, and inline extension factories for custom entrypoints.
└──▷ GET THIS VERSION
$ git clone --branch v0.67.2 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.67.2
└──▷ TRY IT
Layer multiple system prompt fragments at launch — useful when composing role instructions with dynamic context (e.g., rules file + live asset list) without modifying a base config.
$ pi --append-system-prompt "You are a security analyst. Follow OWASP guidelines." --append-system-prompt "Scope: only assess endpoints under /api/v2."
›Supports multiple --append-system-prompt flags in a single invocation, with each value appended to the system prompt separated by double newlines.
›Adds interactive keybinding support for Kitty super-modified shortcuts such as super+k, super+enter, and ctrl+super+k.
›Supports passing inline extension factories to main() for embedded integrations and custom entrypoints.
An open-source AI agent that brings the power of Gemini directly into your terminal.
gemini-cli v0.38.0 adds background memory, context compression, persistent policy approvals, process monitoring, and richer UI controls.
└──▷ GET THIS VERSION
$ git clone --branch v0.38.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:$ git checkout v0.38.0
└──▷ USE IT
Enable non-interactive ADK agent sessions in CI or scripted pipelines where no TTY is present.
json
# In settings.json or gemini config:
{
"experimental": {
"adk": {
"agentSessionNoninteractiveEnabled": true
}
}
}
Check role-specific token usage (e.g. how many tokens the tool-use role consumed) for the current session.
$ /stats
Query an ACP-connected agent for its capabilities and version information without leaving the CLI.
$ /about
›Adds experimental.adk.agentSessionNoninteractiveEnabled config setting for non-interactive ADK agent sessions.
›Introduces a background memory service for automatic skill extraction from sessions.
›Adds a ContextCompressionService to automatically manage and compress long context windows.
›Implements context-aware persistent policy approvals so users don't repeatedly confirm the same tool actions.
›Implements background process monitoring and inspection tools for tracking long-running agent processes.
+15 moreshow less
›Supports default values for environment variables in CLI configuration.
›Adds /about command support for ACP-connected agents.
›Adds /help command for ACP-connected agents.
›Adds role-specific token/usage metrics to the /stats command output.
›Enables web_fetch in plan mode with user confirmation, expanding what agents can do without leaving plan mode.
›Enables compact tool output by default, reducing visual noise in the terminal.
›Adds a scrollbar to the input prompt for long multi-line inputs.
›Introduces minimalist sandbox status labels in the UI.
›Adds TerminalBuffer rendering mode to eliminate UI flicker.
›Enhances tool confirmation UI with an improved selection layout.
›Supports selective topic expansion and click-to-expand for conversation topic headers.
›Displays hook system messages directly in the UI when hooks fire.
›Adds an 'Other' free-text option for yes/no question prompts.
›Scopes subagent workspace directories via AsyncLocalStorage for better isolation.
›Changes default loadingPhrases to off, hiding loading tips by default.
└──▷ BREAKING ON UPGRADE
!The default value of loadingPhrases is now off; users who relied on loading tips/phrases appearing by default will no longer see them without explicitly enabling the setting.
OpenClaw v2026.4.14 adds forward-compat support for GPT-5.4-pro and surfaces human Telegram forum topic names in agent context.
└──▷ GET THIS VERSION
$ git clone --branch v2026.4.14 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:$ git checkout v2026.4.14
›Adds forward-compat support for gpt-5.4-pro (OpenAI Codex family), including Codex pricing/limits and list/status visibility before the upstream catalog catches up.
›Surfaces human Telegram forum topic names in agent context, prompt metadata, and plugin hook metadata by learning names from Telegram forum service messages.
Phoenix 14.5.0 adds span/trace annotation CLI commands, chat metadata on assistant messages, and agent feedback actions.
└──▷ GET THIS VERSION
$ git clone --branch arize-phoenix-v14.5.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:$ git checkout arize-phoenix-v14.5.0
›Adds span and trace annotation commands to the Phoenix CLI.
›Attaches chat metadata to assistant messages in the agent interface.
›Adds feedback actions to assistant messages in the agent interface.
›Enables toggling the assistant agent on or off from Settings.
Phoenix v14.4.0 adds a capability menu and capability registry for the PXI frontend agent tooling.
└──▷ GET THIS VERSION
$ git clone --branch arize-phoenix-v14.4.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:$ git checkout arize-phoenix-v14.4.0
›Adds a capability menu and refactors PXI frontend tool wiring around a capability registry for the agent interface.
Phoenix v14.3.0 adds an assistant agent settings page and re-exports openinference-core from phoenix-otel.
└──▷ GET THIS VERSION
$ git clone --branch arize-phoenix-v14.3.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:$ git checkout arize-phoenix-v14.3.0
›Re-exports openinference-core directly from phoenix-otel, reducing the need for a separate import.
›Adds an assistant agent settings page in the UI for configuring agent behavior.