Verified:GitHub

pm-agent-dev-team

Coordinate multi-agent development with lane plans, ownership, and lock-based coordination.

Updated Aug 23, 2026

Search Tags:#multi-agent #coordination #execution-plan #status-tracking #stakeholder-communication #lane-planning #lock-management

Ready to run

npx skills add https://github.com/akkpol/crafe-flow-nextjs --skill pm-agent-dev-team

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pm-agent-dev-team
Source: https://github.com/akkpol/crafe-flow-nextjs/tree/main/.codex/skills/pm-agent-dev-team
Command: npx skills add https://github.com/akkpol/crafe-flow-nextjs --skill pm-agent-dev-team

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Coordinating multiple coding agents and human team members to deliver features from specs, scope docs, and existing codebases without conflicts.

Core Features & Use Cases

Build context from product docs, active code, and issue context to plan execution.
Partition work into non-overlapping lanes with clear ownership and integration rules.
Protect shared surfaces with lock management and interface contracts before changes.
Produce execution artifacts (execution brief, lane plan, backlog, lock manifest, interface contract, blocker report, status updates) and track delivery quality through defined gates.
Escalate blockers and coordinate cross-lane progress to ensure on-time delivery.

Quick Start

Provide a project brief or feature spec and I will generate a complete multi-lane execution plan with ownership, contracts, and deliverables.

Frequently Asked Questions about pm-agent-dev-team

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

FAQPage Schema

How do I coordinate parallel coding agents to prevent merge conflicts during feature delivery?▼

To prevent merge conflicts during parallel agent coordination, you partition work into non-overlapping lanes with clear ownership and protect shared surfaces using lock management and interface contracts before changes begin.

What execution artifacts are needed for multi-agent development planning?▼

Multi-agent planning requires an execution brief, lane plan, backlog, lock manifest, and interface contracts. These artifacts define ownership boundaries and integration rules to track delivery quality through enforced quality gates.

Can I use lock management to protect shared code surfaces across multiple development agents?▼

Yes, you can use lock management to protect shared code surfaces across multiple agents. Lock manifests and interface contracts establish integration rules, ensuring parallel lanes do not overwrite or conflict on dependent specifications.

How do I generate a multi-lane execution plan from a feature spec?▼

To generate a multi-lane execution plan from a feature spec, provide the project brief to build context from product docs and existing code. The system outputs a complete plan with ownership assignments, lock manifests, and delivery artifacts.

What is the best way to track cross-lane progress and escalate blockers in parallel development?▼

The best way to track cross-lane progress and escalate blockers is through status updates and blocker reports. These execution artifacts monitor cross-lane coordination, ensuring on-time delivery and enforcing quality gates throughout the workflow.

When should I not use lane partitioning for multi-agent software delivery?▼

You should not use lane partitioning for multi-agent delivery when a codebase or feature spec cannot be divided into non-overlapping segments. Without clear ownership boundaries, lock management and interface contracts cannot prevent integration conflicts.