ask-aiLABSCO SUMMARY
Very large, and honest about the one thing that gates it.
Sixty-three tools is a lot to put in front of a model, and the useful framing is the three levels the author describes: stateless code intelligence, embedded LLM agents, then distributed cluster work. The dependency to plan around is stated plainly in the README — without Redis, everything that remembers anything across sessions is off, and you are left with the code-intelligence and AI-call subset.
What it is
A large MCP server that extends a coding assistant with cross-session state, structured cognition and multi-agent orchestration. Redis holds the state between sessions; eight LLM providers can be called from inside it for work your editor's own model does not do. Works with Claude Code, Cursor, Windsurf, VS Code, Zed and any MCP client.
What you get
- 63 tools across 15 categories, with the consolidated ones taking an
actionparameter —task action=create|get|list|update|delete, for instance - Code intelligence that works on meaning as well as text:
find-definition,find-referencesandsemantic-search - Cognition tools that persist what a session learned —
decision-journal,error-pattern,confidence-tracker,intent-tracker,self-critique,smart-handoffandmemory, so the same Redis timeout bug is not re-diagnosed next month - A Kanban layer agents drive directly:
task,task-multi,board,workspace - Multi-model review in one call —
multi-promptfans a prompt out to several models,consensus-promptruns a board-and-CEO vote, andmind-chainhands a question through The Nine Minds (socratic,ontologist,contrarian,simplifierand the rest) so each sees the previous one's output - Autonomous agent runs with
agent-orchestrate, including 2–10 agents in parallel, each with its own orchestration ID;monitorreports status, and because MCP blocks during a tool call, live status is also available over HTTP at/orchestrations - A cluster layer for distributed prompt execution —
cluster-bus-magistraledispatches one prompt across registered clusters under four strategies: first-wins, best-of-n, consensus, fallback-chain
Requirements
No account for the server itself; API keys only for whichever model providers you use — OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY or PERPLEXITY_API_KEY; Ollama needs none. Bun installs it: bun install -g kemdicode-mcp, then claude mcp add kemdicode-mcp -- kemdicode-mcp --stdio. Redis is required for persistence — without it only the stateless tools (code intelligence, AI calls) function; docker run -d -p 6379:6379 redis:alpine is enough. HTTP transport for multi-session use runs with --port 3100. GPL-3.0 licensed.
Setup effort
One command — bun install -g kemdicode-mcp
USE CASES
01
Not re-diagnosing the same bug across projects
Record the failure and its fix once, then match against it the next time the same error shape appears instead of starting the investigation over.
02
Finding your way around an unfamiliar codebase
Locate a symbol's definition and every use, then search by concept rather than exact text, and persist what you learned for the next session.
03
Getting more than one opinion on a risky change
Fan the same review prompt out to several models, or run a board-and-CEO consensus, instead of trusting a single pass.
04
Stress-testing an architecture decision
Chain several Minds so a proposal is challenged, fact-checked and then simplified, each step seeing the one before it.
05
Running several agents at once and watching them
Launch parallel agents with their own orchestration IDs and poll status over HTTP while the run is still blocking the MCP channel.
63 tools
Listed in project docs. Your client may show a different list.
planbuildbrainstormbatchpipelinefind-definitionfind-referencessemantic-searchmulti-promptconsensus-promptenhance-promptmind-chaindecision-journalconfidence-trackermental-modelintent-trackererror-patternself-critiquesmart-handoffcontext-budgetagentagent-commmonitorshared-thoughtsget-shared-contextfeedbacktasktask-multiboardworkspacememorycheckpointinvoke-toolinvoke-batchinvocation-logagent-orchestratesessionthinking-chaingraph-querygraph-find-pathloci-recallsequence-recommendcluster-bus-statuscluster-bus-topologycluster-bus-sendcluster-bus-magistralecluster-bus-flowcluster-bus-routingcluster-bus-inspectcluster-bus-file-readaudit-schedulerclient-samplingclient-elicitclient-rootsenv-infomemory-usageai-configai-modelstool-healthconfigpinghelpBEFORE YOU INSTALL
What to know first
Tools you get
63 tools named in the project's documentation.
Requires
- kemdicode-mcp on PATH
- An API key only for the model provider you use (OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY or PERPLEXITY_API_KEY); Ollama needs none
- Redis for the stateful tools
Where it runs
All 7 documented clients run the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
bun install -g kemdicode-mcpMCP Clients
Claude Code
CLAUDE CODE · TERMINAL
claude mcp add kemdicode-mcp -- kemdicode-mcp --stdioCursor, Windsurf, Roo Code, Other MCP clients
JSON
{
"mcpServers": {
"kemdicode-mcp": {
"command": "kemdicode-mcp",
"args": [
"--stdio"
]
}
}
}The configuration on this page does not set OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY and PERPLEXITY_API_KEY, and you only need the ones for the providers you use. The project's config guide exports them as environment variables (export OPENAI_API_KEY="sk-..."), or you can set a provider key at runtime with its ai-config tool.
VS Code
VS CODE · JSON
{
"servers": {
"kemdicode-mcp": {
"command": "kemdicode-mcp",
"args": [
"--stdio"
]
}
}
}The configuration on this page does not set OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY and PERPLEXITY_API_KEY, and you only need the ones for the providers you use. The project's config guide exports them as environment variables (export OPENAI_API_KEY="sk-..."), or you can set a provider key at runtime with its ai-config tool.
Zed
ZED · JSON
{
"context_servers": {
"kemdicode-mcp": {
"command": {
"path": "kemdicode-mcp",
"args": [
"--stdio"
]
}
}
}
}The configuration on this page does not set OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, DEEPSEEK_API_KEY, OPENROUTER_API_KEY and PERPLEXITY_API_KEY, and you only need the ones for the providers you use. The project's config guide exports them as environment variables (export OPENAI_API_KEY="sk-..."), or you can set a provider key at runtime with its ai-config tool.
