Self Improving Agent
- 975 installs
- 23.5k repo stars
- Updated July 17, 2026
- alirezarezvani/claude-skills
self-improving-agent is a Claude Code skill that promotes auto-generated MEMORY.md entries into permanent project rules, reusable skills, and cleaned documentation for developers who want agents to compound project knowl
About
self-improving-agent is an alirezarezvani claude-skills package that turns Claude Code auto-generated MEMORY.md notes into durable engineering assets. Instead of letting session-specific memory fragments scatter or stale, the skill guides agents to extract repeatable conventions into permanent project rules, publishable skills, and cleaned documentation future sessions can trust. Developers reach for self-improving-agent when Claude Code sessions keep rediscovering the same repository quirks, coding standards, or workflow fixes that belong in version-controlled guidance. The workflow fits long-running repositories where agent memory should evolve into explicit, reviewable artifacts rather than opaque per-session state. It complements manual CLAUDE.md curation by systematizing promotion from ephemeral memory files.
- Analyzes MEMORY.md to surface promotion candidates, stale entries, and consolidation opportunities
- 4 command modes: /si:review, /si:promote, /si:extract, /si:status
- Graduates transient learnings into CLAUDE.md and .claude/rules/
- Extracts recurring solutions into standalone reusable skills
- Delivers memory health dashboard with line counts, topic distribution, and recommendations
Self Improving Agent by the numbers
- 975 all-time installs (skills.sh)
- +11 installs in the week ending Jul 29, 2026 (Skillselion tracking)
- Ranked #1,092 of 16,565 AI & Agent Building skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Jul 31, 2026 (Skillselion catalog sync)
npx skills add https://github.com/alirezarezvani/claude-skills --skill self-improving-agentAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 975 |
|---|---|
| repo stars | ★ 23.5k |
| Security audit | 3 / 3 scanners passed |
| Last updated | July 17, 2026 |
| Repository | alirezarezvani/claude-skills ↗ |
How do you turn agent memory into project rules?
Turn Claude Code's auto-generated MEMORY.md entries into permanent project rules, reusable skills, and cleaned documentation.
Who is it for?
Teams running Claude Code on long-lived repos where session memory should become explicit maintained rules and skills.
Skip if: One-off prototypes without repeated agent sessions or teams that do not use Claude Code MEMORY.md auto-memory features.
When should I use this skill?
Claude Code generates MEMORY.md learnings that should become permanent rules, skills, or cleaned docs for future sessions.
What you get
Permanent project rules, reusable skill files, and cleaned documentation promoted from MEMORY.md entries.
- Permanent project rules
- Reusable skills
- Cleaned documentation
Files
../../../engineering-team/self-improving-agent/skills/self-improving-agent/SKILL.md
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la462k185k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.379k15.8k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.378k92.5k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.375k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k8How it compares
Use self-improving-agent to compound session memory into assets; use cs-note when CodeStable attention.md needs tiny always-on startup facts.
FAQ
What does self-improving-agent do with MEMORY.md?
self-improving-agent reviews Claude Code auto-generated MEMORY.md entries and promotes durable lessons into permanent project rules, reusable skills, and cleaned documentation future agent sessions can load intentionally.
When should teams use self-improving-agent?
Teams should use self-improving-agent on long-running Claude Code projects where the same conventions resurface in memory files and deserve version-controlled rules or skills instead of repeated rediscovery.
Is Self Improving Agent safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.