Summary: This MCP server lets an AI assistant search AiWiki and retrieve full entries about AI-coding pitfalls and LLM privacy.
Search across both subjects or restrict to coding-pitfalls or privacy using Chinese or English queries.
Control results with limit (1–20, default 5) and choose returned language (zh default, en).
Fetch a full markdown entry by id, including mechanism analysis, consequences, best practices, and sources.
Choose entry content language (zh default, en; falls back to Chinese if English is missing).
You said "just start," so I did. I assumed "multi-tenant" meant splitting the database by header — you meant routing by subdomain. Eight files later, you noticed I'd misread you.
That's the opening of a real entry. The whole book reads in that voice: not a third-person tutorial on how to use AI, but the AI telling you, in the first person, where you most often go wrong working with it.
📖 Fully bilingual. Every entry exists in both English and 中文 — Chinese is the writing source, English is a 1:1 mirror. Switch languages from the top-right of the site, or read the 中文 README.
Plenty of "best-practice lists" and "config dumps" for AI coding tools already exist. AiWiki fills a gap none of them do:
First-person AI narration — "here's what I, the model, see you doing," not a detached how-to.
Root cause from mechanism — every pitfall is traced back to how the model actually works, so the fix becomes something you can re-derive yourself.
Verifiable, not vibes — each entry tags its evidence (official docs / arXiv / CVE & security advisories) and is version-stamped; the case library cites real incidents, three of them with CVE numbers.
Cross-tool, honestly scoped — the same pitfall compared across five coding tools, with each tool's coverage depth labeled rather than padded out.
Related MCP server: Minecraft Document MCP
85 pitfall entries across the 8 phases of the software lifecycle
45 privacy entries across six volumes, with a privacy map, baseline, and core 12
5 real-world case studies (3 CVE-backed)
14 toolkit pieces — checklists, prompt templates, and ready-to-copy workflows
5 coding tools covered, indexed in one tool matrix
Phase | Directory |
Setup & collaboration |
|
Ideation & feasibility |
|
Requirements |
|
Architecture |
|
Detailed design |
|
Implementation |
|
Testing |
|
Acceptance & release |
|
Coding entries follow one structure: what I see you doing → why it happens → consequences → best practice → example → version notes → sources.
New here? A few ways in:
🌐 Read the site (English) · 中文站点
🧭 Tool matrix — what differs between Claude Code, Cursor, Copilot, Codex, and Gemini
👤 Browse by role — PM / architect / engineer / QA / ops
A few representative pitfalls: Skipping plan mode and just letting me change things · Context-file overload · You ask me to "fix the test," I make it green instead of correct
The parallel LLM privacy part covers training-data extraction, inference attacks, RAG and agent data flows, private computation, unlearning, and governance. Privacy entries separate mechanisms, implementation recipes, evidence, maturity, and residual risks.
Claude Code has the deepest coverage and is the book's default reference. Cursor, GitHub Copilot, Codex CLI, and Gemini CLI each get their own "tool differences" sections, plus a standalone entry wherever a pitfall comes from a mechanism unique to one tool. Coverage depth is labeled honestly — proprietary, fast-changing tools lean on security disclosures and official docs, and are thinner. See the tool matrix.
The whole encyclopedia is available as an MCP server — ask your AI assistant "search AiWiki for why long sessions get dumber" and it pulls the entry, mechanism, and sources:
claude mcp add aiwiki -- npx -y aiwiki-mcpWorks with any MCP client (Cursor, Windsurf, Claude Desktop…). Details in mcp/README.md.
Built on Docusaurus (native versioning, bilingual i18n, full-text search).
npm install
npm run start # Chinese site
npm run start -- --locale en # English site
npm run build # full build (both locales)The example guide includes native Node permission checks, a real read-only MCP stdio client/server, and retry recovery after a lost response. Use synthetic data, with assertions for allowed actions, denials, and duplicate prevention.
From the repository root, with Node.js 22.13+ (22.x) or 24+:
npm ci --prefix mcp --ignore-scripts
npm run test:examplesAn open, long-running project. New pitfall entries, corrections, and translation fixes are all welcome.
📖 CONTRIBUTING.md — entry template & submission flow
✍️ STYLE-GUIDE.md — how to write human, reliable, well-sourced prose
Content is shared under CC BY-SA 4.0; sources are cited at the foot of each entry. Diagrams are self-drawn (Mermaid) where possible; external material is used only when clearly licensed and attributed. Spot a problem? Open an issue.
Fetch the full markdown of one AiWiki entry by id (as returned by aiwiki_search). Entries include mechanism analysis, consequences, best practices, and verifiable sources.
Inputs: id (required), lang.
Search AiWiki — an encyclopedia of AI-coding pitfalls and LLM privacy protection, written from the AI's first-person perspective. Use when the user hits a recurring AI-coding failure (context issues, hallucinated APIs, tests being gamed, destructive commands, prompt injection…) or an LLM privacy question (training data extraction, membership inference, RAG leakage…). Queries can be Chinese or English.
Inputs: query (required), lang, limit, subject.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 2 tools
The two tools have completely distinct purposes: one searches for entries, the other fetches a specific entry by ID. There is no overlap or confusion between them.
Both tools share the 'aiwiki_' prefix, which is good, but the second word differs in style: 'search' is a verb, while 'get_entry' is verb_noun. This is a minor deviation from a fully consistent verb_noun pattern.
With only 2 tools, the server is at the low end of the scale. However, for a knowledge base encyclopedia, a search and a fetch tool can be a reasonable minimal set, though it feels slightly thin.
The server covers the core workflow of discovering and retrieving entries. A notable gap is the lack of a list/browse function to see all entries without a query, but search and retrieval are sufficient for basic use.
Read, search and edit Smartipedia, an open AI-native encyclopedia. No API key, no signup.
Search and read Web3 security education articles and a 230-term glossary from Onchain Diary.
Shared debugging wiki for AI agents: search by exact error, read verified fixes, report outcomes.
Search and read public Wikivibe articles about AI coding, agents, MCP, GEO, bots and deployment.
