sementis-squad

Coordinates a five-agent squad for research, planning, implementation, and code review workflows.

5|Updated Feb 12, 2026
One-click install
npx skills add https://github.com/PHenrique07/Sementis-IFSP-Pirituba --skill sementis-squad-phenrique07
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sementis-squad
Source: https://github.com/PHenrique07/Sementis-IFSP-Pirituba/tree/main/.github/skills/sementis-squad
Command: npx skills add https://github.com/PHenrique07/Sementis-IFSP-Pirituba --skill sementis-squad-phenrique07

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? This Skill removes ambiguity about which agent should handle a task in the Sementis project by routing work through a defined squad of specialized agents, ensuring multi-file features follow a consistent research-to-review pipeline. ## Core Features & Use Cases - Complexity Triage: SemeIA evaluates each request and chooses between a Fast Path for simple single-file questions and a Full Pipeline for multi-file features or database changes. - Sequential Agent Pipeline: BrinIA researches requirements and proposes technical approaches, CaIA selects the approach and defines JSON payload contracts, GratIA implements the code following project conventions, and XimbeIA reviews for overengineering and rule compliance. - Use Case: When adding a new quiz feature to the Flask and Vanilla JS Sementis platform, the squad researches constraints, plans the API contracts, implements the code in Portuguese naming conventions, and reviews it before delivery. ## Quick Start Use the sementis-squad skill to plan and implement a new daily missions endpoint for the Sementis API.

Frequently Asked Questions about sementis-squad

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

FAQPage Schema
How do I coordinate multiple AI agents for a coding task?▼

Define specialized agent roles and route tasks sequentially through them. This skill uses SemeIA to triage complexity, then BrinIA for research, CaIA for planning, GratIA for implementation, and XimbeIA for review on multi-file changes.

When should I use a full agent pipeline versus a direct answer?▼

Use the Fast Path for simple questions or adjustments touching a single file. Use the Full Pipeline when tasks touch multiple files, introduce new features, or change database rules in models.py.

Can this workflow work with Flask and Vanilla JavaScript projects?▼

Yes, the squad is designed around the Sementis stack of Flask, SQLite, SQLModel, and Vanilla JS. BrinIA identifies these architecture constraints during research, and GratIA respects conventions like Portuguese naming and no extra frontend frameworks.

What happens if implementation requires changing the database models?▼

Changes to models.py must be explicitly highlighted in CaIA's plan before implementation. GratIA is instructed to never alter models.py unless the change was anticipated in the planning stage.

What are the limitations of a sequential agent pipeline?▼

The sequential flow adds overhead for trivial tasks, which is why the Fast Path exists for simple cases. The pipeline also depends on each agent's base document in .github/agents/ being present and accurate.