skills.rest
itsadijmbt
itsadijmbtCOMMUNITY
Explore all skills by this author →
GitHub

self-improvement

Capture agent learnings, corrections, and error context into structured logs.

1|Updated Apr 24, 2026
Search Tags:#debugging#self-improvement#error tracking#agent workflows#hook automation#knowledge promotion#learning logs
One-click install
npx skills add https://github.com/itsadijmbt/SecureMCP-Servers --skill self-improvement-itsadijmbt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/itsadijmbt/SecureMCP-Servers/tree/main/TEST_SERVERS/PORTED_TO_SECUREMCP/alibabacloud-rds-openapi-mcp-server/skill/self-improving-agent
Command: npx skills add https://github.com/itsadijmbt/SecureMCP-Servers --skill self-improvement-itsadijmbt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

It solves the problem of losing valuable debugging context, corrections, and recurring failures by providing a consistent way to capture learnings and errors for later reuse and promotion.

Core Features & Use Cases

  • Capture corrections and knowledge gaps: Log when a user corrects the agent or when understanding is outdated or incomplete.
  • Record failures with context: Log unexpected command/tool/API errors (including integration details) so fixes are possible later.
  • Request and plan missing capabilities: Track feature requests when desired behavior/tools do not exist yet.
  • Promote broadly applicable learnings: Distill durable insights into higher-level workspace memory files for future sessions and workflows.

Quick Start

After a task, ask the agent to log what went wrong or what was learned into .learnings/ using the self-improvement skill format.

Frequently Asked Questions about self-improvement

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

FAQPage Schema
How do I log agent corrections and debugging context for continuous improvement?▼

You can log agent corrections by capturing errors, knowledge gaps, and command failures into structured files within the .learnings/ directory. This allows coding agents to retain debugging context and apply corrections in future sessions.

What is the best way to track API failures and command errors across agent workflows?▼

Tracking API failures involves recording unexpected tool errors and integration details using structured logging formats. This ensures command failures and external API errors are documented for later reuse and troubleshooting.

How do I persist agent knowledge and learning logs beyond a single session?▼

To persist agent knowledge, you promote broadly applicable learnings and durable insights into higher-level workspace memory files. This ensures debugging context and corrections survive beyond a single session.

Can I use hook automation to capture missing capabilities and feature requests?▼

Yes, hook and script support can automatically request and plan missing capabilities. This tracks desired behaviors or tools that do not exist yet within the coding agent workflows.

Does this approach require any specific dependencies to start logging agent errors?▼

No specific dependencies are required to start logging agent errors. The self-improvement process relies on standard scripts, references, and assets to organize learnings without external packages.

When should I not use structured learning logs for agent error tracking?▼

Structured learning logs are not ideal for ephemeral errors that will not recur or when knowledge promotion is unnecessary. Avoid logging transient issues that lack broader debugging context or reusable value.

Related Skills

Explore More Software Engineering Skills→
ArtyMcLabinArtyMcLabin
8

retrospective

Generate a structured retrospective template from session notes after mistakes.

Community
Basic
2lab-ai2lab-ai

learn

Review conversations to extract learnings and durable facts for persistent memory.

Official
Intermediate
technomenschtechnomensch
5

session-wrap

Detect session stopping points and prompt summaries of work progress.

Community
Basic
botlearn-aibotlearn-ai
9

botlearn-selfoptimize

Track errors, corrections, and learnings to markdown memory.

Official
Advanced