codekb-explain

Aggregates symbol summaries, knowledge entries, bug patterns, and call graphs via the codekb_explain MCP tool.

1|Updated Jun 27, 2026
One-click install
npx skills add https://github.com/TE-QuanBZhang/skills-pool --skill codekb-explain-te-quanbzhang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codekb-explain
Source: https://github.com/TE-QuanBZhang/skills-pool/tree/main/codekb/skills/codekb-explain
Command: npx skills add https://github.com/TE-QuanBZhang/skills-pool --skill codekb-explain-te-quanbzhang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Before modifying a function or class, developers often lack context about its intent, historical bugs, and design constraints, leading to regressions and reintroduced defects. This Skill retrieves an aggregated explanation of any symbol so changes are made with full awareness of its behavior and risks. ## Core Features & Use Cases - Aggregated Symbol Explanation: Calls the codekb_explain MCP tool to return a functional summary, associated knowledge entries, known bug patterns, call graph, and coding conventions in one response. - Defect Prevention: Surfaces known_issues sorted by severity so you avoid reintroducing previously fixed bugs when editing code. - Impact Assessment: Reuses CodeGraph to show callers and callees, with a documented fallback to codegraph explore or ripgrep when the graph is unavailable. - Use Case: Before refactoring PaymentService.processRefund, run the Skill to learn it uses a Saga pattern, requires BigDecimal for amounts, and once had a concurrency bug in a shared HashMap. ## Quick Start Ask the AI to explain the symbol PaymentService.processRefund using codekb before modifying it.

Frequently Asked Questions about codekb-explain

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

FAQPage Schema
How do I understand a function before modifying it?

Call the codekb_explain MCP tool with the symbol name, such as Class.method, to get its functional summary, related design decisions, known bug patterns, and call graph. Review the known_issues list before proposing any fix.

What does the codekb_explain MCP tool return?

It returns a summary string, a knowledge array of design decisions and business rules, known_issues sorted by severity, a callgraph of callers and callees, coding conventions, and a code snippet for the requested symbol.

Does codekb_explain require CodeGraph to work?

No, CodeGraph is optional. If CodeGraph is not initialized, the callgraph field returns empty with a _note, and you can fall back to codegraph explore or ripgrep for caller and callee information.

What should I do if the codekb_explain tool is unavailable?

Assemble the same information manually by running codegraph explore for the call graph and reading the files under codekb/knowledge/ for design decisions and bug patterns associated with the symbol.

Why check known bug patterns before editing code?

Known issues reveal why defensive code exists around a symbol. Removing code that guards against a documented bug pattern without confirming the trigger conditions are gone can reintroduce previously fixed defects.