picassioCommunity·11 Agent Skills Included

pi-squad

Parallel multi-agent coding with dependency scheduling and QA rework

Breaks complex coding requests into subtasks and assigns them to specialist agents that run in parallel. Schedules work by dependency order, triggers automatic QA rework loops on failures, and enforces an independent review gate before acceptance. Eliminates manual coordination, vague handoffs, and unverified completion claims with a live progress widget and evidence-based verdicts.

48|3

↗

npx skills add picassio/pi-squad --all -g -y

All Skills in This Repository (11)

Pure Emerald Level Indicators

📦 In Repo

picassio

squad-qa-testing

Automates QA verification with standardized testing frameworks and machine-parsed verdicts.

squad-verification

Verifies task completion through evidence-based checks and structured QA verdicts.

squad-architecture

Plan and document system architecture using YAML and Markdown templates.

squad-collaboration

Break down project tasks, assign roles, and automate QA rework loops.

squad-frontend-dev

Provide frontend development guidelines for React, Tailwind CSS, and accessibility.

squad-debugging

Guide software developers through structured debugging with reproduce-first and hypothesis testing workflows.

squad-code-review

Review source code with two-pass correctness and complexity assessment.

squad-supervisor

Coordinate multi-agent teams and track task progress with QA.

squad-security-audit

Analyze application code for security vulnerabilities and provide remediation guidance.

squad-protocol

Coordinate multi-agent squads with real-time communication and automatic rework loops.

squad-backend-dev

Automate API design, security checks, and testing for RESTful backend development.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install pi-squad?▼

Run `npx skills add picassio/pi-squad --all -g -y` in your terminal to install all skills in this suite globally.

What does pi-squad do?▼

It splits complex coding tasks into subtasks, assigns them to specialist agents like backend, frontend, and QA, and runs them in parallel with dependency-aware scheduling.

How does the QA rework loop work?▼

When a QA agent reports a FAIL verdict, the scheduler automatically creates a fix task for the original agent and a retest task, looping until the work passes or retries are exhausted.

Can I define my own task plan?▼

Yes. You can pass an explicit task list with IDs, agents, and dependencies, or write a strict JSON file spec validated with the bundled validator script.

Do I need coding experience to use pi-squad?▼

No. You describe the goal in plain language and the planner agent creates the task breakdown, while the supervisor handles monitoring and review.