| name | manage-knowledge-base |
| description | Register and organize content in the knowledge base — save web link content, register algorithm training problems, and maintain topic indexes. Use when the user shares a link to register, mentions a LeetCode problem, wants to save study material, or asks to add content to the knowledge base. |
Manage Knowledge Base
Register external content (articles, tutorials, documentation) and algorithm training problems into the structured knowledge-base/ directory.
When to Use
- User shares a web link and asks to register/save its content
- User wants to register an algorithm training problem (LeetCode, etc.)
- User asks to add study material or technical content to the knowledge base
Workflow: Registering Web Link Content
All content goes into knowledge-base/{section-name}/.
Steps
- Fetch the content — Use the browser to open the link and read it
- Evaluate content length and decide format:
- Short content → dump the entire document into a new Markdown file
- Long content → identify relevant sections, create a summary file, and enrich existing files with references to it. If it makes sense, create a dedicated file (full dump or summary)
- Handle images:
- Single image (e.g., infographic) → save alongside the Markdown file in the section folder
- Multiple images → save in a subfolder named after the file
- Place in the correct section — identify the most matching
knowledge-base/ subfolder
- Update the section index — add an entry to the section's
README.md (create one if it doesn't exist; reference knowledge-base/algorithms/README.md as the pattern)
Section folders
| Folder | Content type |
|---|
algorithms/ | Algorithm problems, DSA patterns, competitive programming |
architecture/ | Software architecture patterns, design decisions |
artificial-intelligence/ | AI/ML concepts, tools, frameworks |
courses/ | Course notes and materials |
english-training/ | English practice and language learning |
software-engineering/ | General SWE practices, testing, DevOps |
system-design-interview/ | System design concepts and interview prep |
tech-leads-club/ | Leadership and tech lead resources |
If no existing section fits, create a new one following kebab-case naming.
Workflow: Registering Algorithm Training
Problems are stored at knowledge-base/algorithms/solutions/{source-name}/{problem-name}/problem.md.
Naming
{source-name}: lowercase source (e.g., leet-code, hackerrank)
{problem-name}: {number}-{name-in-kebab-case} (e.g., 209-minimum-size-subarray-sum)
Steps
- Get the problem — the user may provide the problem statement directly or a link to the source
- Create
problem.md with the following structure:
# {Number} - {Problem Name}
**Source:** {Platform}
**Difficulty:** {Easy/Medium/Hard}
**Link:** {URL}
## Problem Statement
{Full problem description}
## Analysis
### Problem Analysis
{Requirements, constraints, math rules, edge cases}
### Possible Solutions
{Approaches with time/space complexity — do NOT write code, just describe the algorithm}
## Topics
{List: Array, Binary Search, Sliding Window, etc.}
- Do NOT write solution code — only analysis and proposed approaches
Maintaining the Topics Index
After creating problem.md, update knowledge-base/algorithms/README.md:
- Add the problem to all relevant topics listed in its Topics section
- Mark the most relevant topic (primary technique in optimal solution) with ⭐
- Maintain alphabetical order of topics
- Use consistent link format:
[{Number} - {Problem Name}](./solutions/{source-name}/{problem-name}/problem.md)
Reference Files
| Purpose | Path |
|---|
| Algorithms index | knowledge-base/algorithms/README.md |
| Example problem | knowledge-base/algorithms/solutions/leet-code/209-minimum-size-subarray-sum/problem.md |