aily-self-improvement

Log agent learnings, errors, and feature requests into markdown files.

2|1|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/GACLove/feishu-aily-skills --skill aily-self-improvement
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aily-self-improvement
Source: https://github.com/GACLove/feishu-aily-skills/tree/main/skills/aily-self-improvement
Command: npx skills add https://github.com/GACLove/feishu-aily-skills --skill aily-self-improvement

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill provides a consistent, auditable way to capture agent learnings, user corrections, tool errors, and feature requests so teams can track recurring issues, prioritize fixes, and promote important knowledge into project memory.

Core Features & Use Cases

  • Structured logging: Templates and ID conventions for Learnings, Errors, and Feature Requests with metadata fields such as priority, status, source, related files, and recurrence tracking.
  • Promotion workflow: Guidance for promoting broadly applicable learnings into AGENTS.md, SOUL.md, TOOLS.md, or MEMORY.md to prevent repeated mistakes and encode team conventions.
  • Detection & review: Triggers for automatic detection of corrections, errors, and feature requests, plus periodic review instructions to resolve, escalate, or promote entries.
  • Use case: After a failed tool execution, log the error with reproduction steps and a suggested fix so the team can assign remediation and avoid repetition.

Quick Start

Log the recent tool failure as an error entry with a concise summary, reproduction steps, suggested fix, and metadata so it is appended to ERRORS.md.

Frequently Asked Questions about aily-self-improvement

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

FAQPage Schema
How do I log agent errors and user corrections in persistent markdown files?

You log agent learnings by capturing tool failures and user corrections into persistent markdown files with standardized metadata, ID generation, reproducible context, and clear promotion paths into project memory for follow-up actions.

What is the best way to track recurring AI agent errors and feature requests?

The best way to track recurring agent errors is to log them with structured metadata, recurrence tracking fields, and status indicators so teams can prioritize fixes and promote broadly applicable learnings into project memory files like AGENTS.md.

How do I promote agent learnings into project memory files like AGENTS.md?

You promote agent learnings into project memory by following a structured workflow that appends broadly applicable knowledge and error fixes into AGENTS.md, SOUL.md, TOOLS.md, or MEMORY.md to prevent repeated mistakes and encode team conventions.

Can I use structured logging to capture knowledge gaps in AI-assisted workflows?

Yes, you can use structured logging to capture knowledge gaps in AI-assisted workflows by applying templates and ID conventions to record missing information, source context, and related files for periodic review, escalation, or promotion into project memory.

Does this agent self-improvement approach work without external dependencies?

Yes, the agent self-improvement approach works without external dependencies by using markdown templates and standardized metadata fields to structure learnings, errors, and feature requests directly within developer environments and AI-assisted workflows.