dynamic-workflows
Dynamic-workflows is a comprehensive orchestration framework that enables teams to coordinate multiple specialized agents across diverse technology stacks. Built for Claude Code adoption at scale, it combines sprint-based workflows, automated browser testing, and intelligent code review systems to streamline development cycles and reduce rework.
Dynamic-workflows enables orchestration of dozens to hundreds of concurrent subagents for large-scale tasks by providing structured patterns for fan-out execution. You can spawn multiple specialized agents simultaneously, coordinate their work through shared context, and aggregate results. The framework handles the complexity of managing concurrent execution, allowing you to focus on task decomposition and result verification rather than low-level concurrency management.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-07-27
Dynamic-workflows enables orchestration of dozens to hundreds of concurrent subagents for large-scale tasks by providing structured patterns for fan-out execution. You can spawn multiple specialized agents simultaneously, coordinate their work through shared context, and aggregate results. The framework handles the complexity of managing concurrent execution, allowing you to focus on task decomposition and result verification rather than low-level concurrency management.
Use it when
- Dynamic-workflows helps you decide between Agent Teams, Dynamic Workflows, and sequential patterns based on your task structure.
- Dynamic-workflows implements fan-out, verification, and pipeline patterns with multiple agents.
Verify before relying
Read SKILL.md below before installing (1 file). Open directory: indexed for reading, not audited.
Similar skills
Install
TheBeardedBearSAS/claude-craft/dynamic-workflows · repository language: Shell
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How can I run hundreds of agents in parallel with dynamic-workflows?
Dynamic-workflows enables orchestration of dozens to hundreds of concurrent subagents for large-scale tasks by providing structured patterns for fan-out execution. You can spawn multiple specialized agents simultaneously, coordinate their work through shared context, and aggregate results. The framework handles the complexity of managing concurrent execution, allowing you to focus on task decomposition and result verification rather than low-level concurrency management.
When should I use Agent Teams versus dynamic-workflows patterns?
Dynamic-workflows helps you decide between Agent Teams, Dynamic Workflows, and sequential patterns based on your task structure. Use Agent Teams for tightly-coordinated groups working on interdependent subtasks; choose Dynamic Workflows when you need flexible, data-driven orchestration across many independent workers; and select sequential patterns for tasks with strict ordering requirements. Dynamic-workflows provides guidance and implementation examples for each approach.
What orchestration patterns does dynamic-workflows support for multi-agent coordination?
Dynamic-workflows implements fan-out, verification, and pipeline patterns with multiple agents. Fan-out distributes work across many agents in parallel; verification patterns run adversarial or independent checks on outputs; and pipeline patterns chain agents sequentially with intermediate filtering or transformation. These patterns can be combined to handle complex workflows like large audits, migrations, and multi-source research tasks.
How do I monitor workflow execution and manage token costs in dynamic-workflows?
Dynamic-workflows provides built-in monitoring and token cost management capabilities. Track workflow execution progress, observe individual agent performance, and analyze token consumption across your orchestrated tasks. This visibility helps you optimize agent allocation, identify bottlenecks, and control costs when scaling beyond single-agent context limits.
Can I save and reuse workflow scripts across different projects?
Yes, dynamic-workflows supports saving and reusing workflow scripts across projects. Define your orchestration patterns, agent configurations, and coordination logic once, then apply them to new tasks and domains. This reusability accelerates development and ensures consistent patterns across your multi-agent implementations.
How many concurrent agents can I run with dynamic-workflows?
Dynamic-workflows is designed to coordinate dozens to hundreds of concurrent subagents, enabling large-scale orchestration far beyond single-agent capabilities. The exact number depends on your task complexity, token budgets, and infrastructure, but the framework provides the patterns and monitoring tools needed to scale responsibly and efficiently.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
Dynamic Workflows — orchestration multi-agents scriptée
Dynamic Workflows (Claude Code 2.1.154+, mot-clé déclencheur ultracode) laissent Claude
écrire un script JavaScript qui orchestre des sous-agents de façon déterministe (boucles,
conditions, fan-out). Le script tourne en arrière-plan ; on suit l'exécution via /workflows.
> ⚠️ Opt-in coûteux : un workflow peut lancer des dizaines à centaines de sous-agents. Ne l'utiliser > que sur demande explicite (mot « ultracode », « use a workflow », « fan out agents ») ou quand le > palier de complexité ci-dessous est franchi.
Quand l'utiliser — les 3 paliers d'orchestration
| Palier | Outil | Concurrence | Quand |
|---|---|---|---|
| 1 | Sub-agent simple |
(truncated - see the full file via the links below)
File tree — 1 file
.claude/skills/dynamic-workflows/SKILL.md
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Orchestrate dozens to hundreds of concurrent subagents for large-scale tasks”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
This skill enables you to search for and access academic theses and research work authored by Allison Fiorentino. It streamlines the process of discovering scholarly contributions across legal and academic domains, making it easier to find relevant publications and research materials.
This skill streamlines the assessment of supplier compliance with GDPR requirements and data processing agreements. It automates the review process for vendor data handling practices, helping legal teams identify regulatory gaps and ensure proper contractual alignment across your supply chain.
This skill automates GDPR compliance assessment for websites, scanning for regulatory adherence across data handling, privacy policies, and user consent mechanisms. It generates detailed audit reports highlighting compliance gaps and recommending corrective actions to meet EU data protection standards.
Lawyerscrib Garry Haas is a specialized legal assistant skill built to handle routine law practice automation. It streamlines document workflows and administrative tasks, enabling legal professionals to focus on higher-value client work and case strategy.
Cqrs teaches the architectural pattern that separates read and write operations into distinct models, enabling systems to scale independently and handle complex business logic more effectively. Explore when CQRS makes sense for your application, how it interacts with event sourcing, and practical implementation strategies across different technology stacks.
This skill helps you spot and strip away telltale markers of machine-generated French content, restoring natural voice and authenticity to your text. Built by humans who understand both AI output and genuine French expression, it works as a practical tool for writers and editors who need to verify content originality.
More skills notaire (MIT) · atomic-tasks (MIT) · Multitenant (MIT) · Worktree Pr (NOASSERTION) · output-workflow-reset (Apache-2.0) · livewire-development (MIT) · Designing Workflow Skills (CC-BY-SA-4.0) · ccw-chain (MIT)