re-feedback

Collects, distills, and archives reverse engineering lessons into skill experience files or GitHub issues.

64|9|Updated Aug 17, 2026
One-click install
npx skills add https://github.com/dslsdzc/rev-skills --skill re-feedback-dslsdzc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: re-feedback
Source: https://github.com/dslsdzc/rev-skills/tree/main/.claude/skills/re-feedback
Command: npx skills add https://github.com/dslsdzc/rev-skills --skill re-feedback-dslsdzc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Reverse engineering insights from analysis sessions, articles, and personal notes are easily lost; this Skill turns them into sanitized, deduplicated, properly categorized experience entries that persist in the skill library or as public GitHub issues. ## Core Features & Use Cases - Three-source collection: Gathers lessons from session retrospectives, article scraping (Kanxue threads and WeChat articles via bin/wxsource.mjs), and manual user notes. - Distillation and sanitization: Formats each lesson as phenomenon/cause/countermeasure and enforces privacy redaction rules that remove company, product, and project identifiers before storage. - Three-tier disposition: Neutrally asks the user to choose between publishing a GitHub issue (via gh CLI), committing locally to a skill's references/experience.md with npm test validation, or discarding. - Use Case: After a malware analysis session, invoke this Skill to distill the pitfalls encountered, classify them to the right skill domain (e.g., re-malware), deduplicate against existing entries, and archive them locally. ## Quick Start Ask the AI to run a session retrospective and distill this analysis session's lessons into the appropriate skill's experience file.

Frequently Asked Questions about re-feedback

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

FAQPage Schema
How do I save lessons learned from a reverse engineering session?▼

Invoke the feedback workflow after an analysis session and it will distill your pitfalls into phenomenon/cause/countermeasure entries. You then choose to publish a GitHub issue, store locally in the target skill's experience.md, or discard.

How to distill Kanxue or WeChat articles into experience entries?▼

Provide the thread ID or article URL and the workflow fetches content via bin/wxsource.mjs, then distills each article individually. Tool promotion posts, truncated articles, and non-technical news are skipped per the intake criteria.

Does the feedback workflow require the GitHub CLI?▼

The gh CLI is only needed for the issue-publishing tier; the workflow checks availability with which gh and gh auth status first. If unavailable, it outputs a markdown draft you can submit manually, and local archiving works without it.

How does it prevent duplicate or privacy-leaking experience entries?▼

Before archiving, it greps the target skill's experience.md for keyword duplicates and skips existing phenomena. Every entry must pass a sanitization self-check removing company names, product names, internal codenames, and identifying detail combinations.

What happens if the lesson's target skill domain is unclear?▼

The workflow uses a decision table mapping topics to skill domains, falls back to grepping all SKILL.md files for keywords, and asks the user rather than guessing when classification remains uncertain.