Heads up This site is currently under heavy development.
Subscribe Get it delivered — the daily firehose, filtered to the tools you run, plus the documentation changes vendors never announce. Compare plans →

The AI Toolchain — issue -088, May 23, 2026

THE AI TOOLCHAIN NO. -088
Tail
THE DAILY RELEASE FIREHOSE
PUBLISHED MAY 23, 2026 · EVERY WEEKDAY
EDITIONS tail grep head diff uniq

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.

VIEW
ISSUE VIEW full issue
Do you prefer this view?
$ tct list   # 2 tools matched
AI & LLM Tooling
◆  AI Coding Agents

GitHub Copilot CLI

Sources Release notes → v1.0.52 NOTES

Copilot CLI v1.0.52 adds deferred tool loading for agents, /compact focus instructions, /usage quota bars, and HTTP/1.1 auto-retry.

└──▷ GET THIS VERSION
$ git clone --branch v1.0.52 https://github.com/github/copilot-cli.git
# already have the repo? check out this version:
$ git checkout v1.0.52
└──▷ TRY IT
Shape what an AI-compacted summary focuses on — useful when a long session covers multiple topics and you want the summary scoped to a specific area.
$ /compact focus on the database migration steps only
Enable deferred tool loading for a custom agent with a large tool list so tool-search discovery works without loading all tools upfront.
yaml
---
name: my-large-agent
deferred-tool-loading: true
---
Resume a saved session from a different working directory, overriding the saved cwd.
$ copilot --continue -C /path/to/other/dir
  • Adds deferred-tool-loading key in agent frontmatter to enable opt-in deferred tool loading for custom agents with large tool lists, unlocking tool-search discovery.
  • Adds optional focus instructions argument to /compact to shape the compaction summary.
  • Adds /usage quota progress bars for session and weekly limits.
  • Migrates legacy nested oauth.clientId and oauth.callbackPort keys in MCP server configs to the supported oauthClientId and auth.redirectPort keys instead of silently dropping them.
  • Adds -C <dir> flag override so sessions resume in their saved working directory, with flags whose values are relative paths (e.g. --attachment, --log-dir) resolving from the saved cwd.
+8 moreshow less
  • Status line command now supports plain shell commands in addition to executable script paths.
  • Requests that time out due to an HTTP/2 upload stall automatically retry over HTTP/1.1.
  • Reasoning tokens now display as a parenthetical on the output token count in the token usage summary.
  • Context window tier selection (default ~200K vs 1M tokens) is now enforced end-to-end, constraining compaction, truncation, and token display.
  • Non-interactive subcommands (plugin list, mcp list, help, version) no longer consume stdin.
  • General-purpose subagents use GPT-5.4 or GPT-5.5 when available.
  • /restart and /update preserve the current session ID after restarting.
  • Adds vertical scrollbar with mouse drag support to the main conversation view.
Was this useful?

Earendil Works Pi

Sources Release notes → v0.75.5 NOTES

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

Pi v0.75.5 adds adaptive thinking for custom Anthropic-compatible providers and cleaner read tool UI.

└──▷ GET THIS VERSION
$ git clone --branch v0.75.5 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.75.5
└──▷ USE IT
Enable adaptive thinking on a custom Anthropic-compatible provider so Claude reasons more deeply on complex tasks.
yaml
# In your custom provider model config:
compat:
  forceAdaptiveThinking: true
  • Adds compat.forceAdaptiveThinking option to custom Anthropic-compatible model configs to enable adaptive-thinking Claude behavior on third-party providers.
  • Collapsed read tool cards now show only the read line by default; press Ctrl+O to expand full file content.
  • Adds a standard unified patch to edit tool result details for SDK consumers.
Was this useful?
my-toolchain — 0 tools
paste an install list to detect your tools

A brew list, a Brewfile, requirements.txt, a Dockerfile — or just the product names, free-form. Nothing leaves your browser.

    browse all tools →