Gives your AI coding agent a confidence-weighted memory for repeated patterns. Each `observe` call records a tool sequence, user preference, fix, or combo; repeated patterns auto-promote from raw to mature to rule to universal across projects. The MCP surface exposes `suggest` for mature guidance, `consolidate` for threshold promotion, `detect_chains` for sequential discovery, and `export_platform` to push rules back into Claude, Cursor, Windsurf, Codex, or CLAUDE.md files. Storage is local SQLite WAL at `~/.instinct/instinct.db`. Reach for it when your agent applies the same correction three times and you want automatic reinforcement instead of manual rule editing. Supports stdio, SSE, and streamable HTTP transports.
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent