By dayvidpham
Aura protocol — 12-phase audit-trail workflow for structured software development with agent coordination, beads task tracking, and vertical slice implementation
Specification writer and implementation designer
Master orchestrator for full 12-phase workflow
End-user alignment reviewer for plans and code
Task coordinator, spawns workers, manages parallel execution
Test writer using BDD, DI, mocks, and Vitest patterns
Pass ratified PROPOSAL-N to supervisor for implementation.
Create PROPOSAL-N Beads task with full specification.
Add `aura:p6-plan:s6-ratify` label to accepted PROPOSAL-N after consensus and UAT.
Send PROPOSAL-N task to reviewers for feedback.
Specification writer and implementation designer
Executes bash commands
Hook triggers when Bash tool is used
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Multi-agent orchestration toolkit for the Aura Protocol. Provides CLI tools for launching Claude agents in isolated worktrees and tmux sessions, role-based slash commands, and a Home Manager module for declarative distribution.
aura-plugins solves two problems in multi-agent Claude workflows:
Isolation. Parallel agents editing the same repo need separate working
trees and branches to avoid conflicts. aura-swarm creates per-epic git
worktrees with automatic Beads context gathering.
Coordination. Agents need structured roles (architect, supervisor,
worker, reviewer) with different toolsets and instructions. Both CLI tools
inject role-specific system prompts from skills/*/SKILL.md files.
All inter-agent coordination flows through Beads task status and comments — there is no separate messaging system.
# 1. Add the aura-plugins marketplace
claude plugin marketplace add dayvidpham/aura-plugins
# 2. Install the aura plugin (user-wide)
claude plugin install aura@aura-plugins
# Or install for a specific project only
claude plugin install aura@aura-plugins --scope project
# Validate (optional)
claude plugin validate /path/to/aura-plugins
After installing, restart Claude Code. Skills are auto-discovered and invocable as /aura:<skill-name> (e.g. /aura:epoch, /aura:user-request).
# Update to latest
claude plugin update aura@aura-plugins
# Local development (from a checkout, no install needed)
claude plugin marketplace add /path/to/aura-plugins
claude plugin install aura@aura-plugins
OpenCode natively reads skills/*/SKILL.md. Clone into your project's .claude/ or configure via OpenCode's plugin system.
Add as a flake input:
# flake.nix
{
inputs.aura-plugins.url = "github:dayvidpham/aura-plugins";
}
Then reference the packages:
# In your system or home-manager config:
aura-plugins.packages.${system}.aura-swarm # unified agent orchestration
aura-plugins.packages.${system}.aura-parallel # deprecated wrapper
aura-plugins.packages.${system}.default # all tools
The flake exports a Home Manager module that syncs commands, agents, and
protocol docs into ~/.claude/:
# home.nix
{
imports = [ aura-plugins.homeManagerModules.aura-config-sync ];
CUSTOM.programs.aura-config-sync = {
packages.enable = true; # install CLI tools
commands.enable = true; # install slash commands
commands.roles = { # toggle per-role
architect = true;
supervisor = true;
worker = true;
reviewer = true;
epoch = true;
};
# commands.enableAll = true; # shorthand for all roles
agents.enable = true; # install custom agents
protocol.enable = false; # install skills/protocol/ docs (off by default)
};
}
Core commands (aura:status, aura:impl:*, aura:user:*) are always installed when
commands.enable = true, regardless of role selection.
Both scripts are standalone Python 3.10+ with no external dependencies:
git clone https://github.com/dayvidpham/aura-plugins
cd aura-plugins
chmod +x bin/aura-swarm bin/aura-release
# Add bin/ to PATH or symlink into a PATH directory
Both tools check for these on startup and exit with an error if any are missing:
| Tool | Purpose |
|---|---|
git | Worktree and branch management |
tmux | Agent session hosting |
bd | Beads issue tracking |
claude | Claude CLI (agent runtime) |
Python 3.10+ (stdlib only, no pip dependencies).
Creates an isolated git worktree for a Beads epic, gathers task context from the dependency graph, and launches a single Claude instance in a tmux session. The agent uses Claude's Agent Teams internally to spawn workers on sub-branches.
main
└── epic/<epic-id> aura-swarm creates this branch + worktree
├── agent/<task-id-1> Claude Agent Teams creates sub-branches
├── agent/<task-id-2>
└── agent/<task-id-3>
aura-swarm <subcommand> [options]
start — Launch an epic sessionCreates (or reuses) a worktree, discovers tasks and URD from the Beads dependency graph, builds a context-rich prompt, and launches Claude in a detached tmux session.
aura-swarm start --epic <epic-id> [options] [task-id ...]
npx claudepluginhub dayvidpham/aura-plugins --plugin auraPasture — a 12-phase protocol for structured multi-agent software development, delivered as Claude Code skills and agents (supervisor, worker, reviewer, architect, epoch + phase sub-skills) generated from the pasture Go codegen. Beads task tracking and vertical-slice TDD implementation.
Security filter that blocks Claude Code from accessing sensitive files (SSH keys, credentials, env files) using the opencode-security-filter Python CLI
Multi-agent orchestration framework for Claude Code, Gemini CLI, and Codex CLI — 19 agents, 14 skills, 16 commands, quality gates, TDD enforcement
This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.
High-intelligence Claude Code copilot with deep code reasoning, evidence-driven planning, orchestration-first execution, model routing, context budgeting, CI/CD integration, enterprise security, plugin development, prompt engineering, performance profiling, agent teams, channels (event-driven autonomy with CI webhook, mobile approval relay, Discord bridge, and fakechat dev profile), interactive tutorials, LSP integration, security-hardened hook script library, MCP Prompts coverage, common workflow packs, runtime selection guide, computer-use patterns, checkpointing, scheduled-task blueprints, repo bootstrap scanner, hook policy engine (8 installable packs), layered memory deployment, role-based subagent packs (implementer, debugger, migration-lead, dependency-auditor, release-coordinator), 5 agent-team topology kits, autonomy operating mode (4 profiles + 3 gates), and a queryable 15-tool MCP documentation server with autonomy advisor.
Multi-agent /workflow development pipeline (planner → plan-review → coder → code-review) with typed handoff contracts, lifecycle hooks, and MCP servers.
11 agents, 35 skills, 18 commands, 9 hooks — spec-driven multi-agent orchestration for Claude Code, with optional cross-device semantic memory.
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities