Skip to content

Anthropic

A Claude Code plugin that refuses to write the code for you

Written by an AI model from the linked sources and checked against them by a second model. How we work

All about: Claude Code versions and changes →
Image: github.com

Learning Mode, posted to Show HN, is a guided-learning skill for Claude Code and Codex. It first reads your intent: quick answer, learning session, or plain result. In learning mode it states a single goal, checks the minimum prerequisite knowledge, then runs a loop of predict, attempt, feedback, repeat.

The learning goal stays with the student. If the goal is modeling form state in React, you write that model yourself; the assistant only produces an empty test scaffold and the surrounding project files.

It installs from the melsayedx/learning-mode plugin marketplace, and you need to start a new conversation after installing. Progress journals are opt-in and land in ~/learning-journals/date-topic/journal.md.

Anthropic

Anthropic’s CEO signs the Pacing the Frontier petition

Anthropic’s CEO, several co-founders, and senior staff signed Pacing the Frontier, a petition backing a deliberate pace of AI development. Anthropic ties the stance to its June research on recursive self-improvement and says the industry broadly agrees.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.