convoy

Coordinate batched work across rigs with stage-driven dispatch and safety guards.

17.5k|1.6k|Updated Dec 16, 2025
One-click install
npx skills add https://github.com/gastownhall/gastown --skill convoy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: convoy
Source: https://github.com/gastownhall/gastown/tree/main/docs/skills/convoy
Command: npx skills add https://github.com/gastownhall/gastown --skill convoy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Gas Town's convoy system tracks batches of work across rigs, enabling safe, event-driven feeding and stage-launch workflows with robust safety guards.

Core Features & Use Cases

  • Event-driven feeder and stranded scan for automatic task dispatch
  • DAG-based wave planning, block/dependency checks, and cross-rig coordination
  • Re-staging, launch, and lifecycle management with clear status transitions

Quick Start

Stage a new convoy for an epic and review the resulting wave plan before launching.

Frequently Asked Questions about convoy

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

FAQPage Schema
How do I coordinate batched work across multiple rigs with dependency management?

You coordinate batched work across rigs by using a stage-driven dispatch system that applies DAG-based wave planning, dependency checks, and safety guards to ensure safe task execution.

What is event-driven feeding in a batch workflow orchestration system?

Event-driven feeding in batch workflow orchestration is a mechanism that automatically dispatches tasks by triggering stranded-convoy scans and stage transitions based on real-time work events.

How do I plan and compute dependency-based waves before launching a batch?

You plan dependency-based waves by staging a new convoy, which computes a wave plan using a DAG to sequence tasks, followed by reviewing block and dependency checks before launch.

Can I use safety guards to prevent unsafe launches in a batch orchestration pipeline?

Yes, safety guards and fail-safe checks are applied throughout the lifecycle to prevent unsafe launches, validating status transitions and cross-rig coordination before execution.

What is the best way to handle stranded convoys in a stage-driven dispatch system?

The best way to handle stranded convoys is through an automated stranded-convoy detection scan that identifies stalled batches and re-triggers event-driven feeding to resume the workflow.

How do I re-stage and manage the lifecycle of a batch after a failed launch?

You re-stage a failed batch by resetting its status transitions within the convoy lifecycle, allowing you to adjust the wave plan and re-attempt a guarded launch across the rigs.