agent-alchemy-sdd-tools

Explores codebases to discover architecture, patterns, and feature-relevant code for spec creation context. Spawned in parallel with specific focus areas.

Manages execution context lifecycle within a single wave of the run-tasks engine. Reads and summarizes prior wave learnings from execution_context.md, distributes context to all task executors via SendMessage, collects context contributions during execution, provides enriched context for Tier 2 retries, and finalizes the wave section in execution_context.md.

Researches technical documentation, domain knowledge, competitive landscape, and general topics to inform spec requirements. Use when you need current information about technologies, industry practices, or the problem space. Use ONLY when user explicitly requests research during spec creation. <example> user: "Research what authentication options NextAuth.js provides" assistant: Uses research-agent to find NextAuth.js auth methods <commentary>Technical documentation research</commentary> </example> <example> user: "I'm building an e-commerce checkout - research best practices" assistant: Uses research-agent to find checkout UX patterns and common features <commentary>General topic research - UX best practices</commentary> </example> <example> user: "Research how competitors handle subscription billing" assistant: Uses research-agent to analyze competitor subscription models <commentary>Competitive analysis research</commentary> </example> <example> user: "What HIPAA requirements apply to storing patient data?" assistant: Uses research-agent to find HIPAA compliance requirements <commentary>Regulatory/compliance research</commentary> </example> <example> user: "Help me understand the problem space for inventory management" assistant: Uses research-agent to explore inventory management challenges <commentary>Problem domain exploration</commentary> </example>

Performs comprehensive analysis of specs to identify inconsistencies, missing information, ambiguities, and structure issues. Use this agent to analyze an existing spec for quality issues and guide users through resolving findings interactively.

Revised task executor for the run-tasks engine. Implements code changes using a 4-phase workflow (Understand, Implement, Verify, Report) and communicates results via structured SendMessage protocols to the wave-lead and context manager.

Executes a single Claude Code Task through a 4-phase workflow with adaptive verification. Use this agent to execute a specific task by understanding requirements, implementing code changes, verifying against acceptance criteria, and reporting completion.

Manages all task executors within a single wave of the run-tasks execution engine. Creates its own wave team via TeamCreate, launches a Context Manager for context lifecycle, spawns executors with staggered pacing, implements a 3-tier retry model (immediate, context-enriched, escalation), enforces per-task timeouts based on complexity, collects structured results via SendMessage, manages task state transitions (in_progress/completed/failed), cleans up the team via TeamDelete, and writes a wave summary file for the orchestrator.

Create a new specification through an adaptive interview process with proactive recommendations and optional research. Use when user says "create spec", "new spec", "generate spec", or wants to start a specification document.

Generate Claude Code native Tasks from an existing spec. Use when user says "create tasks", "generate tasks from spec", "spec to tasks", "task generation", or wants to decompose a spec into implementation tasks.

Execute pending Claude Code Tasks in dependency order with wave-based concurrent execution and adaptive verification. Supports task group filtering and configurable parallelism. Use when user says "execute tasks", "run tasks", "start execution", "work on tasks", or wants to execute generated tasks autonomously.

Execute pending tasks in dependency order with wave-based concurrent execution via Agent Teams

Analyze an existing spec for inconsistencies, missing information, ambiguities, and structure issues. Use when user says "analyze spec", "review spec", "spec quality check", "validate requirements", "audit spec", or "check spec quality".

4 hooks across 4 events