subagent-driven-development_skill

This skill coordinates multiple specialized agents to orchestrate complex, multi-agent development workflows and ensure efficient collaboration across tasks.
  • Python

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill chunkytortoise/enterprisehub --skill subagent-driven-development

  • SKILL.md11.6 KB

Overview

This skill provides patterns and tools for coordinating multiple specialized agents to execute complex development workflows. It focuses on orchestrating autonomous agents with clear roles, dependency management, parallel execution, and quality gates. Use it to run distributed development teams, enforce review stages, and recover from agent failures.

How this skill works

Register specialized agents into a central Workflow Orchestrator and declare tasks with agent_type, priority, dependencies, and input_data. The orchestrator resolves dependencies, schedules tasks (including parallel fan-out), retries on failures, and routes work to fallback agents when needed. Monitoring APIs expose workflow progress, agent health, and task-level status for observability and intervention.

When to use it

  • Coordinating multiple specialized agents for a feature or product pipeline
  • Orchestrating parallel development work with dependency constraints
  • Implementing automated quality gates, security scans, or production approvals
  • Managing autonomous development teams with fallbacks and retry strategies
  • Running iterative refinement loops for optimizations or reviews

Best practices

  • Define clear, non-overlapping responsibilities for each agent type
  • Explicitly model dependencies and validate the task graph before execution
  • Register fallback agents and set sensible retry and timeout policies
  • Add quality gate tasks for production-critical flows and security reviews
  • Monitor workflow and agent health continuously and expose actionable alerts

Example use cases

  • Sequential pipeline: Architect → Developer → Tester → Reviewer → Quality Gate for a feature release
  • Parallel fan-out: split frontend, backend, and database work after a single architecture task
  • Quality gate workflow: include security and quality-gate agents before production deploy
  • Real-estate platform: orchestrate architecture, parallel API/model development, integration testing, and production approval
  • Emergency fast-track: invoke security and developer agents with elevated priority for critical fixes

FAQ

Common agent types include ARCHITECT, DEVELOPER, TESTER, REVIEWER, SECURITY, PERFORMANCE, QUALITY_GATE, and COORDINATOR.

How are task dependencies enforced?

The orchestrator validates and resolves dependencies before scheduling; tasks run only after their dependencies complete successfully or meet configured fallback rules.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational