How do I install attune-ai?

attune-ai is a local plugin. Install it using PyPI package: pip install attune-ai and add the generated configuration snippet to your AI app's MCP config file. Then restart your AI app.

What AI apps work with attune-ai?

attune-ai uses the Model Context Protocol (MCP) and works with any MCP-compatible AI app, including Claude, ChatGPT / Codex, Gemini, Copilot, Cursor, and more.

Back to Browse

attune-ai MCP Server

by Patrick Roebuck

Developer ToolsModerate5.2Local

Free

AI-powered developer workflows for Claude Code — smart routing, multi-agent orchestration, zero conf

About

Attune AI is a developer workflow framework that supercharges Claude Code with intelligent task orchestration. It provides 19 built-in workflows for code review, security auditing, test generation, documentation, and more — all accessible through natural language commands. Features include multi-tier cost optimization (automatically routing between cheap and premium models), multi-agent team coordination, guided wizards for complex tasks, and an MCP server for seamless Claude Code integration. Install with pip, configure nothing, and start shipping better code faster.

Security Report

5.2

Moderate5.2Moderate Risk

The Attune AI MCP server is a developer tools framework with comprehensive functionality for AI-powered workflows. The server properly implements authentication through API keys and environment variables, includes reasonable security measures like path validation, and provides permissions appropriate for its developer tools purpose. Some minor code quality and documentation concerns exist but do not significantly impact the overall security posture. Package verification found 1 issue (1 critical, 0 high severity).

4 files analyzed · 5 issues found

Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.

Permissions Required

This plugin requests these system permissions. Most are normal for its category.

File System Read

Reads files on your machine. Normal for tools that analyze or process local data.

File System Write

Writes or modifies files on your machine. Check that this is expected for the tool.

HTTP Network Access

Connects to external APIs or services over the internet.

env_vars

Check that this permission is expected for this type of plugin.

Shell Command Execution

Runs commands on your machine. Be cautious — only use if you trust this plugin.

process_spawn

Check that this permission is expected for this type of plugin.

How to Install

Add this to your MCP configuration file:

{
  "mcpServers": {
    "attune-ai": {
      "args": [
        "attune-redis"
      ],
      "command": "uvx"
    }
  }
}

Getting Started

Once installed, try these example prompts and explore these capabilities:

1I just installed attune-ai. Help me set it up and explore what it can do.
21. Run `attune doctor` to check my environment
32. Show me the available workflows with `attune workflow list`
43. Run a quick security audit on my current project
54. Explain how the /attune command works for natural language routing
6I want to understand the Socratic discovery flow — how attune asks questions before executing, and how to use the command hubs (/dev, /testing, /workflows, /docs, /plan).

Documentation

View on GitHub

From the project's GitHub README.

Attune AI

The 21st century help system for developer tools.

Heads up: attune-help and attune-author have moved to their own marketplace at Smart-AI-Memory/attune-docs. If you installed them via this marketplace previously, see the migration guide below. New users should add the attune-docs marketplace directly.

Static docs rot. READMEs go stale the moment you merge. Help pages don't know if you're a beginner or an expert. Nobody maintains them — and it shows.

Attune AI is a different approach. Documentation is authored once as templates, rendered at runtime with audience awareness, maintained automatically by AI agents, and learned from based on how people actually use it. The result is a living knowledge base that stays accurate, adapts to who's reading, and improves over time — without anyone manually updating markdown files.

The same system powers 18 multi-agent workflows, 14 auto-triggering skills, and 36 MCP tools — all of which double as the authoring and assistance toolkit for building and maintaining knowledge bases at scale.

How It Works

1. Authored as Templates

633 templates across 11 types — errors, warnings, tips, references, tasks, FAQs, notes, quickstarts, concepts, troubleshooting, and comparisons. Each template has structured frontmatter (tags, related links, audience hints) and a markdown body. Templates are the source of truth; rendered output is ephemeral.

2. Rendered at Runtime

Help adapts to the reader. Progressive depth escalates across template types as you ask again:

First ask   → concept   (what is this?)
Second ask  → task      (how do I use it?)
Third ask   → reference (show me the details)

Audience adaptation adjusts verbosity and framing for Claude Code users, CLI users, and marketplace readers — from the same source template.

Precursor warnings surface relevant errors and warnings before you hit them, based on the file you're editing.

3. Maintained by AI

A 5-phase maintenance workflow detects stale templates, prioritizes by usage feedback, regenerates via batch API, rebuilds cross-links, and validates the result — all without manual intervention.

detect → map → regenerate → rebuild → validate

Templates that help people more get maintained first. Templates nobody reads get deprioritized. The knowledge base optimizes itself.

4. Learned from Usage

Every template lookup is tracked. Feedback ratings adjust template confidence scores. Usage telemetry weights priorities so the maintenance workflow focuses on what matters. The help system gets better the more you use it.

The Toolkit

The help system doesn't just contain knowledge — it comes with tools to build, maintain, and deliver it. These same tools power attune-ai's own 633 templates, proving the approach works at scale.


18 Multi-Agent Workflows	Code review, security audit, test gen, release prep — specialist teams of 2-6 Claude subagents that also serve as knowledge-authoring pipelines
36 MCP Tools	Every workflow exposed as a native Claude Code tool via Model Context Protocol, including `help_lookup` (4 modes) and `help_maintain` (auto-regeneration)
14 Auto-Triggering Skills	Say "review my code" and Claude picks the right skill — each skill integrates contextual help from the template engine
Portable Security Hooks	PreToolUse guard blocks eval/exec and path traversal; PostToolUse auto-formats Python
Socratic Discovery	Workflows ask questions before executing, not the other way around

Get Started in 60 Seconds

Plugin (works standalone)

claude plugin marketplace add Smart-AI-Memory/attune-ai
claude plugin install attune-ai@attune-ai

Then say "what can attune do?" in Claude Code. That's it.

Add Python Package (unlocks CLI + MCP)

pip install 'attune-ai[developer]'

What Each Layer Adds

Capability	Plugin only	Plugin + pip
14 auto-triggering skills	Yes	Yes
Security hooks	Yes	Yes
Prompt-based analysis	Yes	Yes
36 MCP tools	--	Yes
`attune` CLI	--	Yes
Multi-agent workflows	--	Yes
Help system maintenance	--	Yes
CI/CD automation	--	Yes

The plugin works standalone — skills guide Claude through analysis using your existing subscription, with no additional costs. Add the Python package when you want MCP tool execution, CLI automation, help system maintenance, or multi-agent orchestration.

Note: The Python package's CLI and MCP tools use the Anthropic API directly, which requires an API key and incurs usage-based charges. See API Mode for details.

Cheat Sheet

All 14 skills trigger automatically from natural language — just describe what you need:

Input	What Happens
"what can attune do?"	Auto-triggers `attune-hub` — guided discovery
"build this feature from scratch"	Auto-triggers `spec` — brainstorm, plan, execute
"review my code"	Auto-triggers `code-quality` skill
"scan for vulnerabilities"	Auto-triggers `security-audit` skill
"generate tests for src/"	Auto-triggers `smart-test` skill
"fix failing tests"	Auto-triggers `fix-test` skill
"predict bugs"	Auto-triggers `bug-predict` skill
"generate docs"	Auto-triggers `doc-gen` skill
"plan this feature"	Auto-triggers `planning` skill
"refactor this module"	Auto-triggers `refactor-plan` skill
"prepare a release"	Auto-triggers `release-prep` skill
"tell me more"	Auto-triggers `coach` — progressive depth help
"run all workflows"	Auto-triggers `workflow-orchestration` skill

Skills run using your Claude subscription — no API key needed, no additional charges.

Why Attune?

	Attune AI	Static Docs	Agent Frameworks	Coding CLIs
Self-maintaining docs	AI-maintained, usage-weighted	Manual, rots immediately	None	None
Progressive depth	concept → task → reference	One-size-fits-all	None	None
Audience adaptation	Adapts per reader	Write multiple versions	None	None
Ready-to-use workflows	18 built-in	None	Build from scratch	None
Multi-agent teams	2-6 agents per workflow	None	Yes	No
MCP integration	36 native tools	None	No	No
Portable security hooks	PreToolUse + PostToolUse	None	No	No

Workflows

Every workflow runs as a multi-agent team. Each agent reads your code with Read, Glob, and Grep tools and reports findings to an orchestrator that synthesizes a unified result.

Workflow	Agents	What It Does
code-review	security, quality, perf, architect	4-perspective code review
security-audit	vuln-scanner, secret-detector, auth-reviewer, remediation	Finds vulnerabilities and generates fix plans
deep-review	security, quality, test-gap	Multi-pass deep analysis
perf-audit	complexity, bottleneck, optimization	Identifies bottlenecks and O(n^2) patterns
bug-predict	pattern-scanner, risk-correlator, prevention	Predicts likely failure points
health-check	dynamic team (2-6)	Project health across tests, deps, lint, CI, docs, security
test-gen	identifier, designer, writer	Writes pytest code for untested functions
test-audit	coverage, gap-analyzer, planner	Audits coverage and prioritizes gaps
doc-gen	outline, content, polish	Generates documentation from source
doc-audit	staleness, accuracy, gap-finder	Finds stale docs and drift
dependency-check	inventory, update-advisor	Audits outdated packages and advisories
refactor-plan	debt-scanner, impact, plan-generator	Plans large-scale refactors
simplify-code	complexity, simplification, safety	Proposes simplifications with safety review
release-prep	health, security, changelog, assessor	Go/no-go readiness check
doc-orchestrator	inventory, outline, content, polish	Full-project documentation
secure-release	security, health, dep-auditor, gater	Release pipeline with risk scoring
research-synthesis	summarizer, pattern-analyst, writer	Multi-source research synthesis

MCP Tools

36 tools organized into 4 categories:

Workflow (20)

security_audit code_review bug_predict performance_audit refactor_plan simplify_code deep_review test_generation test_audit test_gen_parallel doc_gen doc_audit doc_orchestrator release_prep health_check dependency_check secure_release research_synthesis analyze_batch analyze_image

Help (5)

help_lookup help_init help_status help_update help_maintain

Memory (4)

memory_store memory_retrieve memory_search memory_forget

Utility (7)

auth_status auth_recommend telemetry_stats context_get context_set attune_get_level attune_set_level

Installation Options

# Recommended (agents, memory)
pip install 'attune-ai[developer]'

# Minimal (CLI + workflows only)
pip install attune-ai

# All features
pip install 'attune-ai[all]'

# Development (contributing)
git clone https://github.com/Smart-AI-Memory/attune-ai.git
cd attune-ai && pip install -e '.[dev]'

API Mode

The plugin's skills use your Claude subscription at no extra cost. The Python package's CLI and MCP tools work differently — they spawn Agent SDK subagents that make direct Anthropic API calls, which require an API key and incur usage-based charges.

export ANTHROPIC_API_KEY="sk-ant-..."     # Required
export REDIS_URL="redis://localhost:6379"  # Optional

Model Routing

Each subagent is assigned a model based on task complexity to balance cost and quality:

Model	Agents	Rationale
Opus	security, vuln, architect	Deep reasoning
Sonnet	quality, plan, research	Balanced analysis
Haiku	complexity, lint, coverage	Fast scanning

export ATTUNE_AGENT_MODEL_SECURITY=sonnet  # Save cost
export ATTUNE_AGENT_MODEL_DEFAULT=opus     # Max quality

Budget Controls

Every CLI/MCP workflow enforces a budget cap:

Depth	Budget	Use Case
`quick`	$0.50	Fast checks
`standard`	$2.00	Normal analysis (default)
`deep`	$5.00	Thorough multi-pass review

export ATTUNE_MAX_BUDGET_USD=10.0  # Override

Security

Path traversal protection on all file operations (CWE-22)
Memory ownership checks (created_by validation)
MCP rate limiting (60 calls/min per tool)
Hook import restriction (attune.* modules only)
PreToolUse security guard (blocks eval/exec, path traversal)
Prompt input sanitization (backticks, control chars, truncation)
PII scrubbing in telemetry
Automated security scanning (CodeQL, bandit, detect-secrets)

See SECURITY.md for vulnerability reporting and full security details.

Migration

attune-help and attune-author have moved to their own marketplace at Smart-AI-Memory/attune-docs. If you previously installed either of them via the attune-ai marketplace, move your installation with the three commands below.

Add the new marketplace:

/plugin marketplace add Smart-AI-Memory/attune-docs

Uninstall from the old marketplace:

/plugin uninstall attune-help@attune-ai
/plugin uninstall attune-author@attune-ai

Install from the new marketplace:

/plugin install attune-help@attune-docs
/plugin install attune-author@attune-docs

New users: add Smart-AI-Memory/attune-docs directly — no migration steps needed.

Links

Apache License 2.0 — Free and open source.

If you find Attune useful, give it a star — it helps others discover the project.

Acknowledgments

Special thanks to:

Anthropic — For Claude AI, the Model Context Protocol, and the Agent SDK patterns that shaped v5.0.0
Boris Cherny — Creator of Claude Code, whose workflow posts validated Attune's approach to plan-first execution and multi-agent orchestration
Affaan Mustafa — For battle-tested Claude Code configurations that inspired our hook system

View Full Acknowledgements

Built by Patrick Roebuck using Claude Code.