Claude Code v2.1.170: Claude Fable 5 Model Access and VS Code Session Fix

Claude Code v2.1.170 introduces Claude Fable 5, a new Mythos-class model that Anthropic has deemed safe for general use. According to the release, Fable's capabilities exceed those of any model previously made generally available. The update also fixes a session-saving bug affecting VS Code integrated terminal users.
Key Changes
- Claude Fable 5 (Mythos-class): New high-capability model, safer for general deployment. Capabilities surpass all prior generally available models. Requires updating to v2.1.170 for access.
- Session Transcript Fix: Sessions launched from the VS Code integrated terminal (or any shell inheriting Claude Code environment variables) were not saving transcripts and failed to appear in
--resumeoutput. This is now fixed.
Getting the Update
Update Claude Code to version 2.1.170 to access Claude Fable 5. If you're using VS Code's integrated terminal, the session fix is especially relevant — your transcripts will now persist correctly.
Who This Affects
All Claude Code users interested in evaluating new models, and anyone relying on --resume for session continuity, particularly those using VS Code or custom shell environments that propagate Claude Code env vars.
📖 Read the full source: GitHub Claude-Code
👀 See Also

GitHub Copilot Moves to Usage-Based Billing by Token Consumption, Replacing Premium Requests on June 1, 2026
GitHub Copilot transitions from premium request units to token-based GitHub AI Credits, with plan prices unchanged. All paid plans include monthly credits equal to subscription cost; additional usage billed at API rates.

OpenClaw 2026.4.29 Breaks Setups: CPU Spikes, Tool Restrictions, and Fixes
OpenClaw 2026.4.29 introduces CPU spikes from active-run steering, restricted tool profiles breaking exec/fs commands, and stricter group chat handling. Rollback or apply targeted fixes.

When Everyone Has AI but the Company Still Learns Nothing: The Messy Middle of Enterprise AI Adoption
Ethan Mollick's framework shows that individual AI productivity gains don't automatically become organizational learning. The article explores why companies are stuck in a 'messy middle' where AI use is uneven, hidden, and disconnected from shared knowledge.

Claude Code v2.1.119: Config Persistence, GitLab/Bitbucket PR Support, and Dozens of Bugfixes
Claude Code v2.1.119 persists /config settings to ~/.claude/settings.json, adds --from-pr support for GitLab MRs and Bitbucket PRs, and fixes over 25 bugs including CRLF paste, MCP OAuth, and auto-mode conflicts.