solo-agent-dispatch

Dispatch role-based sub-agents to execute parallel or sequential project tasks.

Updated Apr 23, 2026
One-click install
npx skills add https://github.com/cjrain-12505614/solo-forge-marketplace --skill solo-agent-dispatch-cjrain-12505614
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: solo-agent-dispatch
Source: https://github.com/cjrain-12505614/solo-forge-marketplace/tree/main/solo-forge-v0.6.0/skills/solo-agent-dispatch
Command: npx skills add https://github.com/cjrain-12505614/solo-forge-marketplace --skill solo-agent-dispatch-cjrain-12505614

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Coordinating complex solo development projects requires splitting work across specialized roles, but manually delegating tasks to sub-agents without clear structure leads to inconsistent outputs, ignored decisions, and duplicated effort. ## Core Features & Use Cases - Flexible Team Composition: Select agents from a pool of 10 roles (researcher, planner, designer, architect, FE/BE developer, DevOps, QA, data analyst, tech writer) based on project scale from small to large. - Standardized Dispatch Templates: Use role-specific prompt templates that embed project background, the latest decisions from decision_log.md, success criteria, and output locations. - Parallel and Sequential Orchestration: Run independent tasks concurrently (e.g., research plus planning) or chain dependent phases (requirements to design to implementation) with decision-log synchronization points. - Use Case: When starting a new feature, dispatch a researcher and a planner in parallel to produce a tech-stack survey in docs/01_참고자료/ and a requirements spec in docs/02_기획/, then collect results and report consolidated outcomes to the PM. ## Quick Start Ask the assistant to dispatch a researcher agent to investigate the technology stack and save the findings to the docs reference folder.

Frequently Asked Questions about solo-agent-dispatch

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I dispatch multiple agents in parallel for a project?▼

Identify independent tasks such as research and planning, then dispatch each agent simultaneously with its own role-specific prompt template. Each agent writes to a separate output folder like docs/01_참고자료/ or docs/02_기획/, so parallel work does not conflict.

What should an agent dispatch prompt include?▼

Include the agent role and name, project background, the latest five entries from decision_log.md, a single clear mission, success criteria, and the exact output path and filename. Always state that the agent should execute immediately without asking for plan confirmation.

How do I choose team size for a solo development project?▼

Match team size to project scale: small projects with 5-15 features need 2-3 agents, medium projects with 15-50 features need 4-5 agents, and large projects with 50+ features use the full role pool including architect, DevOps, and QA.

Why does a dispatched agent ignore previous decisions?▼

Agents ignore decisions when the prompt omits decision_log.md content. Fix this by citing the specific decision number, stating it was already finalized, and re-including the relevant log entries in the dispatch prompt before redeploying.

When should agents run sequentially instead of in parallel?▼

Use sequential dispatch when tasks have dependencies, such as requirements before architecture design or implementation before testing. Synchronize between steps through decision_log.md entries and completed artifacts from the prior agent.