Understand this story
Speaking at the AI Engineer Code Summit, Lydia Hallie, who leads developer relations for Claude Code at Anthropic, said that for the Claude 5.5 series models, many of the Claude Code usage practices accumulated over the past year no longer apply. She recommends switching the model to Opus 5.5, changing permissions to auto mode, and trimming the emphasis instructions in CLAUDE.md that were written for older models. She also offered safety and quality fallback measures: deny rules and PreToolUse hooks take effect before the classifier, sandboxes and cloud sessions limit side effects, hooks use exit code 2 to feed failures back to the model, and CI uniformly covers the pull requests opened by each session. There is currently no further development regarding this assessment.
AI-generated from public reports
Latest development · Report
Claude Code Updates Projects: Coordinator More Proactive, Supports /btw and /forkDevelopments
1 confirmed developments
Lydia Hallie says old Claude Code best practices no longer apply
Source:Shao Meng
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 14
Claude DevsFirst-partySelected
Claude Code Updates Projects: Coordinator More Proactive, Supports /btw and /fork
Claude Code's Projects has received several updates: the coordinator within a project has become more proactive, answering questions in the project chat and syncing the status of parallel threads. The new /btw opens a separate thread for a task that comes up ad hoc, while /fork branches a new thread off the current one to try a different approach; threads can be given reminders, and when the time comes Claude brings them up back in the project chat. A single thread can be shared as a read-only link with users who are logged in to Claude; threads that have accessed the local machine cannot be shared. A row of recently opened, updated, or pinned Artifacts has been added at the top of the thread panel. Projects is currently in public beta on Pro and Max; users who do not yet have it can join the waitlist.
Nov 13
Shao Meng
Anthropic's Lydia Hallie on Reassessing Claude Code Best Practices
Lydia Hallie, who leads developer relations for Claude Code at Anthropic, argued in a talk at the AI Engineer Code Summit that, for the Claude 5.5 series models, many Claude Code usage patterns accumulated over the past year no longer apply. She recommended switching the model to Opus 5.5, changing permissions to auto mode, and trimming the emphasis instructions written for older models in CLAUDE.md. She also offered safety and quality fallback measures: deny rules and PreToolUse hooks take effect before the classifier, sandboxes and cloud sessions limit side effects, hooks use exit code 2 to feed failures back to the model, and CI uniformly covers the pull requests opened by each session.