SkillsCat
SkillsCat
Trending
Docs
LingmaFuture

LingmaFuture

@LingmaFuture

GitHub
1 Skills
1 Total Stars
May 2026 Joined

Public Skills

self-improvement

by LingmaFuture

"Capture durable lessons from debugging, user corrections, missing capabilities, and repeated workflow friction so future sessions avoid the same mistakes. Hybrid design: actual-self-improvement execution core + self-improving-compound HOT/WARM/COLD memory tiers + legacy promotion/hook guidance. Use immediately when a non-obvious failure is diagnosed, before the final reply after a workaround succeeds, when the user corrects or updates the agent, when a project/tool convention is discovered, when a missing capability is requested, or when a solved issue should be promoted into shared memory. Also use to audit whether a lesson was captured. Do not use for trivial typos, expected failures, straightforward retries, or one-off noise with no reusable lesson."

Processing 1 22d ago
SkillsCat
SkillsCat

An open platform for discovering, sharing, and installing AI agent skills.

Discover

  • Trending
  • Recently Added
  • Top Rated
  • Categories

Resources

  • Docs
  • Claude Code Docs
  • Claude Code GitHub
  • Cursor

Legal

  • Privacy Policy
  • Terms of Service

© 2026 SkillsCat. Open source under AGPL-3.0.

Docs · Privacy · Terms