use in your agent
Let your agent look up the term.
The same 323 terms, in the places your agent already reads. When you describe a problem in plain words, the agent can find the engineering term, tell you its name, and follow its check.
Complete worked examples
11 entries include a bug, a short prompt, before and after code, a runnable check, limits, and primary sources. The skill reference and JSON include these examples; MCP’s get_term returns the context and download links.
Claude Code skill
The skill loads only when a request describes a problem without naming it. It keeps a short index in context and reads the full entry on demand. Run this once:
mkdir -p ~/.claude/skills/engineering-terms/references && curl -fsSL https://vibe-coding-101.brighteng.org/skill/SKILL.md -o ~/.claude/skills/engineering-terms/SKILL.md && curl -fsSL https://vibe-coding-101.brighteng.org/skill/references/terms.md -o ~/.claude/skills/engineering-terms/references/terms.md
Files: SKILL.md · references/terms.md. Put them in .claude/skills/engineering-terms/ inside a repository to share the skill with your team.
MCP server
A hosted, read-only server with three tools: find_term takes a problem in your own words, get_term returns one entry, and list_terms lists them. Nothing to install.
Claude Code
claude mcp add --transport http vibe-coding-101 https://vibe-coding-101.brighteng.org/mcp
Cursor, in .cursor/mcp.json
{
"mcpServers": {
"vibe-coding-101": {
"url": "https://vibe-coding-101.brighteng.org/mcp"
}
}
}Other clients: add a Streamable HTTP server at https://vibe-coding-101.brighteng.org/mcp.
Packs for CLAUDE.md and AGENTS.md
A pack is a Markdown section with one line per term: when to use it, what to do, and what to avoid. Append the packs that match your stack. The full pack is long, so prefer the skill or MCP server for everything.
curl -fsSL https://vibe-coding-101.brighteng.org/packs/frontend.md >> AGENTS.md
- all terms323 terms · /packs/all.md
- development workflow18 terms · /packs/practice.md
- agents and AI16 terms · /packs/agent.md
- code design22 terms · /packs/eng.md
- frontend and browser state46 terms · /packs/frontend.md
- APIs and data77 terms · /packs/backend.md
- performance12 terms · /packs/game.md
- security13 terms · /packs/security.md
- testing12 terms · /packs/feedback.md
- concurrency27 terms · /packs/concurrency.md
- native apps25 terms · /packs/mobile.md
- delivery and operations43 terms · /packs/ship.md
- web interface12 terms · /packs/ux.md
JSON
Every term with its prompt, problem phrases, and related terms: /terms.json. Plain text for language models: /llms-full.txt.