Mindkeg MCP
未认领A persistent memory MCP server for AI coding agents — stores, searches, and retrieves atomic learnings so agents retain knowledge across sessions.
安装
npx mindkeg-mcp init工具（8 个）
complete_run
Record a completed work session
get_context
Retrieve relevant knowledge — session primer, task-scoped context, or semantic search (replaces getcontext, getrelevantcontext, searchlearnings)
list_scopes
List repositories and workspaces with counts (replaces listrepositories, listworkspaces)
query
List knowledge by type — decisions, findings, gotchas, or runs (replaces getdecisions, getopenfindings, getgotchas, getrunhistory)
relate_learnings
Create typed relationships between learnings
resolve
Close out a decision or finding (replaces supersededecision, resolvefinding)
store
Save knowledge — learning, decision, finding, or gotcha (replaces storelearning, storedecision, storefinding, storegotcha)
update
Modify/manage knowledge — update, deprecate, flagstale, delete, or merge (replaces updatelearning, deprecatelearning, flagstale, deletelearning, mergelearnings)