skills.rest
ShubhamManachekar
ShubhamManachekarCOMMUNITY
Explore all skills by this author →
GitHub

self-improving-agent

Promote proven session patterns into GeoSupply governance rules and SKILL.md entries.

Updated Mar 5, 2026
Search Tags:#code-quality#memory-management#ai-handshake#architecture-drift#promotion-patterns#geosupply
One-click install
npx skills add https://github.com/ShubhamManachekar/GeoSupply --skill self-improving-agent-shubhammanachekar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: self-improving-agent
Source: https://github.com/ShubhamManachekar/GeoSupply/tree/main/.agent/skills/self-improving-agent
Command: npx skills add https://github.com/ShubhamManachekar/GeoSupply --skill self-improving-agent-shubhammanachekar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

GeoSupply teams face architecture drift and inconsistent application of session learnings across AI handoffs. This skill curates memories and promotes proven patterns into enforceable project rules, CLAUDE.md, and domain SKILL.md to maintain alignment and reduce rework.

Core Features & Use Cases

  • Automates memory curation and rule propagation across memory tiers (CLAUDE.md, DEVELOPMENT_TRAIL.md, and .agent/skills/*/SKILL.md).
  • Supports commands for scanning, promoting, extracting new skills, and syncing handoff trails.
  • Enables anti-drift governance for GeoSupply FA workflows and developer handoffs with auditable promotion paths.

Quick Start

Instruct the agent to run a memory review and promote verified patterns to CLAUDE.md or geosupply-dev/SKILL.md.

Frequently Asked Questions about self-improving-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent architecture drift during AI handoffs?▼

To prevent architecture drift during AI handoffs, promote proven session patterns into enforceable governance rules across CLAUDE.md, DEVELOPMENT_TRAIL.md, and SKILL.md memory tiers.

How does memory curation work for live development sessions?▼

Memory curation works by scanning live development sessions, extracting verified patterns, and applying them across a three-tier architecture to maintain project alignment and reduce rework.

How do I promote session learnings into project rules?▼

You promote session learnings by instructing the agent to run a memory review, which scans verified patterns and propagates them into CLAUDE.md or domain SKILL.md files as enforceable rules.

Do I need YAML frontmatter for SKILL.md files to prevent architecture drift?▼

Yes, YAML frontmatter with name and description is required in SKILL.md files to properly enforce governance rules and prevent architecture drift during AI handoffs.

What is the best way to sync handoff trails across memory tiers?▼

The best way to sync handoff trails is using dedicated scanning and promotion commands that propagate verified patterns across the three-tier memory architecture for auditable governance.

What limitations exist when extracting new skills from live sessions?▼

Extracting new skills requires verified patterns from live sessions and supports optional script, reference, or asset resources for on-demand use, limiting promotion to proven development outcomes only.

Related Skills

Explore More Software Engineering Skills→
MiniMax-AIMiniMax-AI
2.9k

skill-creator

Generate Skill Units with YAML frontmatter, Markdown body, and bundled resources.

Official
Intermediate
fredrick84823fredrick84823
2

create-handoff

Generate YAML-frontmatter handoff documents to transfer work context between agent sessions.

Community
Intermediate
New-dev0New-dev0
1

mem0-codex

Store and retrieve coding project memories using mem0 MCP tools.

Community
Intermediate
ddaanetddaanet

codify

Consolidate session learnings into permanent, searchable documentation.

Community
Advanced