skills.rest
fyrsmithlabs
fyrsmithlabsOFFICIAL
Explore all skills by this author →
GitHub

self-reflection

Analyze agent behavior memories to identify failure patterns and output remediation recommendations.

Updated Jan 7, 2026
Search Tags:#remediation#self-reflection#ReasoningBank#CLAUDE.md#agent-behavior#memory-analysis#contextd
One-click install
npx skills add https://github.com/fyrsmithlabs/marketplace --skill self-reflection-fyrsmithlabs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: self-reflection
Source: https://github.com/fyrsmithlabs/marketplace/tree/main/plugins/contextd/skills/self-reflection
Command: npx skills add https://github.com/fyrsmithlabs/marketplace --skill self-reflection-fyrsmithlabs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams systematically review agent behavior patterns, surface actionable findings, and improve CLAUDE.md style and ReasoningBank health, reducing recurring mistakes and misalignments.

Core Features & Use Cases

  • Behavioral self-review loop: analyzes past interactions to identify rationalized-skip, overclaimed, and ignored-instruction patterns.
  • Surface findings to users: generate clear remediation recommendations and improve documentation.
  • Use cases: after failures, during evaluation, or before major releases to surface issues and remediation opportunities.

Quick Start

Run the self-reflection process after a critical interaction to surface findings and suggest CLAUDE.md improvements.

Frequently Asked Questions about self-reflection

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

FAQPage Schema
How do I analyze agent behavior memories to identify repeated failure patterns?▼

Analyzing agent behavior memories involves running a self-reflection loop that reviews past interactions to detect rationalized-skip, overclaimed, and ignored-instruction patterns, then surfacing actionable findings for remediation.

When should I run an agent behavior self-review to improve CLAUDE.md health?▼

Run an agent behavior self-review after a series of failures, during model evaluation, or before major releases to surface behavioral issues and generate structured remediation recommendations for your CLAUDE.md.

Does the agent memory analysis process require a specific MCP server for cross-session access?▼

Yes, the self-reflection process uses the contextd MCP server to access and analyze cross-session memory, enabling comprehensive identification of behavioral issues and remediation opportunities across past interactions.

How do I fix rationalized-skip and ignored-instruction patterns in my AI agent?▼

Fix rationalized-skip and ignored-instruction patterns by applying a guided self-reflection process that analyzes past behaviors and outputs structured remediation recommendations to improve your CLAUDE.md and ReasoningBank health.

What is the best way to generate remediation recommendations before a major model release?▼

The best way to generate remediation recommendations before a release is to run a behavioral self-review loop that examines agent memories to surface issues and suggest CLAUDE.md documentation improvements.

Related Skills

Explore More Data & Analytics Skills→
fyrsmithlabsfyrsmithlabs

using-contextd

Initialize contextd MCP sessions for cross-session memory and semantic search.

Official
Advanced
anam-organam-org
118

claude-improve-config

Analyze sessions to propose CLAUDE.md, hook, and skill improvements.

Official
Basic
AbdullahMalik17AbdullahMalik17
9

evaluation

Evaluate agent performance with weighted multi-dimensional rubric scores.

Community
Advanced
Snowflake-LabsSnowflake-Labs
12

read-ors-configuration

A concise summary of your roleplay character: name,</name> personality, and/or behavior, formatted as a single string sentence. Typically不超过50字，此处省略

Official
Intermediate