Claude.ai Auto-Compact Feature Still Not Working Despite January Fix Announcement
By
nurimamedov
Slightly gummy. The crust never quite set.
Summary
A user reports that the auto-compact feature in Claude.ai (both web and desktop versions) is not working despite being marked as fixed by Anthropic on January 15. When the context window fills up, instead of automatically compacting to continue conversations, messages either get bounced back to the input box without error or occasionally trigger a "limit reached" error. This occurs even when the context shouldn't be near the 200k token limit. The user has followed proper bug reporting procedures by searching existing issues and confirming they're using the latest version.
Key quotes
· 4 pulledAuto-compact doesn't work. Anthropic flagged this as fixed on January 15, but the issue persists.
When context window fills up, instead of auto-compacting and continuing the conversation, one of two things happens: Messages get bounced back to the input box with no error (most common).
This happens even when the context being fed the chat shouldn't be anywhere near the 200k token limit.
Auto-compact should trigger automatically when approaching context limits, allowing the conversation to continue.
You might also wanna read
Claude Code Login Bug: OAuth Timeout Error on Windows Prevents Google Authentication
A user reports a bug where Claude Code fails to log in on Windows with an OAuth timeout error when attempting to authenticate via Google. Th
github.com·1mo ago
Multiple @redhat-cloud-services npm packages compromised in supply chain attack
Multiple npm packages under the @redhat-cloud-services scope have been compromised with malicious releases. The affected packages include @r
github.com·1h ago
Why Agentic Coding and AI-Driven Development May Lead to Cognitive Atrophy
The article argues that the current hype around "Agentic Coding" and Spec Driven Development (SDD)—where AI handles all coding while humans
larsfaye.com·1h agoWhy Agentic Coding and AI-Driven Development May Lead to Cognitive Atrophy
The article argues that the current hype around "Agentic Coding" and Spec Driven Development (SDD)—where AI handles all coding while humans
larsfaye.com·1h ago
A 5-Step Method for Building Reusable AI Agent Skills from Repeatable Processes
A practical guide on building reusable AI agent skills by turning repeatable processes into automation. The article outlines a 5-step method
dsebastien.net·1h ago
xAI Releases Grok Build 0.1 Coding Model to Developers via Public API Beta
xAI has released Grok Build 0.1, its fastest coding model, to developers via the xAI API in public beta. Previously limited to paying subscr
devops.com·2h ago