Guides
Every team knows more than its wiki says: why something was built the way it was, which edge cases exist, what was agreed with whom. These guides show how to capture that knowledge so people and AI agents can find it. Your wiki stays what it is.
- Team memory via MCP: what it is and when it pays offA team memory gives AI agents via MCP the knowledge missing between the rules file and the wiki. How to tell you need one, and what matters.
- Confluence via MCP: what your AI finds there and what it does notVia MCP, your AI finds in Confluence what was written down as a page. What is learned day to day is rarely there. What belongs where, and a two-week test.
- Giving Claude Code a memory for your teamHow Claude Code learns the why behind your code: decisions, edge cases and agreements that live only in people's heads, as a shared memory via MCP.
- CLAUDE.md or knowledge base: when is one file enough?A CLAUDE.md is enough for rules that apply to every task. When team knowledge outgrows it and a searchable knowledge base via MCP helps.
- Wiki and tacit knowledge: what belongs in the wiki and what does notDocumentation stays in the wiki. Edge cases, agreements and the why in people's heads need a lighter place beside it that your AI searches too.
- Getting support knowledge out of people's headsSupport edge cases and workarounds are often written down nowhere. How to capture them so every shift and your AI agent find them instead of asking.
- Onboarding: passing on the unwritten knowledgeHow to pass on to new team members what no handbook says: edge cases, agreements and the why, findable instead of only in the heads of the experienced.