Best Focus AI Skills & MCP Servers

AI copilot for TouchDesigner with live MCP control — 106 tools, focus + locations, hint injection, component notes, POPx inspection, knowledge corpus, project lifecycle control, technique memory, typed patch sessions, agent activity log, one-tool self-upd

Full project audit workflow — code quality review, security scan, architecture analysis, and metrics report.

Autopilot workflow — autonomous task execution with reasoning, context management, and shell commands.

Batch execution patterns — run multiple operations sequentially, in parallel, or as a pipeline.

Agent registry — register AI agents with capabilities, list available agents, query capabilities.

Knowledge base — index documents, search by content, fetch entries, purge stale data, rank by relevance.

List directory contents — entries, tree, glob, find.

Change impact analysis — given a file or symbol, find what files are affected by a change.

Project-level understanding without reading all files — detect framework, language, conventions, and architecture.

Full-text search with TF-IDF ranking — index files, search with relevance scoring, suggest completions.

Format data into various output formats — pretty JSON, YAML, markdown, ASCII tables.

In-memory file cache — avoid redundant disk reads with mtime-based invalidation.

Composite brick — orchestrates smartread, cache, compress, tokenbudget, and overview to deliver optimal context within a token budget.

Text search and replace across files — regex search, multi-file replace, grouped results.

Knowledge graph export — generate HTML, Mermaid, GraphML, Cypher, Obsidian, and wiki formats.

Package version intelligence — latest version, diff, changelog, audit across npm/pypi/github/gitlab.

API route detection — scan projects for Express, Fastify, Next.js, SvelteKit route definitions.

Token budget management — estimate and optimize token usage for AI agents.

Write files — create, overwrite, append.

Developer tools — loads shell, sandbox, batch for command execution and scripting.

Read files — full, head, tail, line range, encoding detection.

Pack files into a compressed context bundle — reduce token usage by extracting only signatures, maps, or full content.

Task dispatch queue — send tasks, manage queue, cancel, check status for multi-step workflows.

Graph clustering — detect communities, explain groupings, identify architecture layers.