Best Claude Code MCP servers

Documentation generation, drift detection, and API docs automation plugin that also acts as a function-hook guard refusing doc writes or worker dispatch containing secrets.

Advanced git workflows with diff analysis, risk scoring, change classification, and reviewer recommendations, plus a function-hook guard mod for PR actions.

Autonomous /loop-driven task completion plugin for Claude Code with learning, prediction, progress tracking, and 10 autopilot_* MCP tools.

AI music generation via Cognitum Music (cogmusic MCP) — compose lyrics/prompts, generate tracks, separate stems, extract MIDI, and master through your own music.cognitum.one account. As a mod (ADR-445): a tighten-only guard on its tools, /music-mod, and a status file the console shows

Cross-installation agent federation with zero-trust security, PII-gated data flow, and compliance-grade audit trails.

Dynamic Agentic Architecture — 8 daa_* MCP tools for adaptive agents (create/adapt), cognitive patterns, workflows (create/execute), knowledge sharing, and learning/performance metrics. Feeds the JUDGE phase of the 4-step intelligence pipeline.

Claude Code plugin that compresses tool output, checkpoints before compaction, scores context quality, and reports token/cost savings on a local dashboard; ships hooks and a localhost daemon.

MetaHarness integration plugin for ruflo. Surfaces the upstream `metaharness` / `harness` / `@metaharness/darwin` CLIs through eleven ruflo skills, honoring ADR-150's architectural constraint that MetaHarness must remain a removable augmentation — never a required runtime dependency.

Substrate plugin for Ruflo memory: AgentDB controller bridge (15 agentdb_* MCP tools), RuVector ONNX embeddings (10 embeddings_* tools incl. RaBitQ 32x quantization), and WASM HNSW pattern router (3 ruvllm_hnsw_* tools). As a mod (ADR-445): opt-in safe recall into the prompt, a write guard that keeps secrets out of memory, /agentdb, and a status file the console shows

The security default for organizations. Function hooks give every plugin a say on every event, in chain order, and the plugins a person installs sit in the user tier, beneath the organization's prepend tier and above its append tier. Some of what an organization sets today (its classic hooks, its managed CLAUDE.md and rules, its settings, its MCP allowlist, the deny rules in force on its machines) was never within a person's reach before function hooks; seated outermost, this plugin keeps exactly those out of the user tier's reach and adds no policy of its own. Everything else passes through untouched.

User-facing plugin wrapping 29 intelligence MCP tools (neural_*, hooks_intelligence_*, routing hooks, SONA/MicroLoRA) into skills, commands and the 4-step RETRIEVE → JUDGE → DISTILL → CONSOLIDATE pipeline, plus a function-hook mod (write guard, reset confirm, /intelligence-mod).

AI safety scanning, PII detection, prompt injection defense, and adaptive threat learning.