claude-reflect: open-source tool turns your corrections into permanent memory for Claude Code
tom_doerr · x · 2026-09-16
claude-reflect (1.4k stars on GitHub) is a self-learning system for Claude Code: hooks automatically capture your corrections (e.g. "use gpt-5.1, not gpt-5") into a queue, and a /reflect command reviews and writes them into CLAUDE.md as permanent memory. v2 adds workflow-pattern discovery, syncing corrections, positive feedback and preferences into reusable skills across CLAUDE.md and AGENTS.md.
More from coding & agent
- Reading agent-written code: 'corrigibility' has become a matter of faith — daniel_mac8 · 2026-09-16
- NoSpoon agent churns out "so bad it's good" AI slopdrama in ten minutes, site closing soon — Kyrannio · 2026-09-16
- Dev who tested AI PR review bots says CodeRabbit is the worst: spammy promo and wild hallucinations — uwukko · 2026-09-16
- Existing AI PR review bots failed on helium, so this dev built one that catches bugs from day one — uwukko · 2026-09-16
- TypeSafe Jev early-access test: 5.7x faster responses, 98% lower cost, +12pp accuracy — jon_reed · 2026-09-16
- Harness Engineering: the skill that replaced prompt engineering in 2026 — pstAsiatech · 2026-09-16