skills.rest
JoernStoehler
JoernStoehlerCOMMUNITY
Explore all skills by this author →
GitHub

feedback-processing

Generalize corrective feedback into error classes and record them in MEMORY.md.

Updated Feb 6, 2026
Search Tags:#feedback#memory#review#generalization#interactive-session#error-class
One-click install
npx skills add https://github.com/JoernStoehler/msc-math --skill feedback-processing
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feedback-processing
Source: https://github.com/JoernStoehler/msc-math/tree/main/.claude/skills/feedback-processing
Command: npx skills add https://github.com/JoernStoehler/msc-math --skill feedback-processing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Processing feedback during interactive sessions requires generalizing a single correction into a broader error class, identifying all related instances, and maintaining a durable memory of the error for future sessions.

Core Features & Use Cases

  • Generalize isolated feedback into a reusable error class.
  • Scan the current session for other instances of the same error class and dispatch subagents to address them.
  • Record the error class and context in MEMORY.md to enable cross-session recognition and learning.

Quick Start

Initiate the feedback-processing workflow when Jörn provides corrections, then run the generalization loop to update instances, abstract the error class, and archive the results in MEMORY.md.

Frequently Asked Questions about feedback-processing

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

FAQPage Schema
How do I generalize isolated feedback into a reusable error class during an interactive session?▼

To generalize feedback, you isolate the reasoning error from a single corrected instance, abstract it into a universal error class, scan for related instances, and record the class in MEMORY.md for cross-session recognition.

What is the best way to scan an interactive session for recurring reasoning errors after a correction?▼

The best way to scan for recurring reasoning errors is to apply a feedback-processing workflow that abstracts the correction into an error class and dispatches subagents to find and fix all related instances in the current session.

How does recording error classes in MEMORY.md improve cross-session feedback processing?▼

Recording error classes in MEMORY.md improves cross-session feedback processing by maintaining a durable memory of generalized corrections, enabling the system to recognize and avoid the same reasoning errors in future interactive sessions.

Can I dispatch subagents to fix multiple instances of the same error class automatically?▼

Yes, you can dispatch subagents to fix multiple instances automatically once the feedback-processing workflow abstracts a corrected instance into a universal error class and scans the interactive session for related occurrences.

What are the limitations of generalizing feedback without a structured tracking workflow?▼

Without a structured tracking workflow, generalizing feedback lacks re-review prompts and memory updates, meaning related instances of the error class may be missed and the correction will not achieve cross-session coverage.

Related Skills

Explore More Content & Communication Skills→
veniceaiveniceai
130

venice-responses

Process user inputs via Venice's Responses API into typed reasoning, messages, tool calls, and web searches.

Official
Advanced
TrossitecTrossitec
1

self-correction-loop

Capture user corrections and persist generalized rules in MEMORY.md.

Official
Intermediate
tarotmansatarotmansa

compound-engineering

Scan agent sessions and consolidate learnings into MEMORY.md files.

Community
Intermediate
shekerkammashekerkamma

log-correction

Log analyst mistakes and fixes in structured YAML correction logs.

Community
Intermediate