using-project-skills

Enforces checking the project skills catalog and invoking matching skills before responding to tasks.

Updated Jun 11, 2026
One-click install
npx skills add https://github.com/coferlandia/coferlandia-skills --skill using-project-skills-coferlandia
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: using-project-skills
Source: https://github.com/coferlandia/coferlandia-skills/tree/main/skills/meta/using-project-skills
Command: npx skills add https://github.com/coferlandia/coferlandia-skills --skill using-project-skills-coferlandia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents often skip available project skills because a task feels too small, too obvious, or too urgent, causing reusable workflows and operational contracts to go unused. This Skill acts as a mandatory gate that forces a skills-catalog check before any response, file exploration, or clarifying question. ## Core Features & Use Cases - Mandatory Discovery Gate: Requires checking the nearest skills/INDEX.md or equivalent catalog and invoking any matching skill before acting on a task. - Skill Priority Rules: Defines ordering when multiple skills apply, such as running process skills like software-development before domain skills. - Rationalization Detection: Lists common red-flag thoughts ("it's just a quick question", "I already know this") that signal an improper skip of the skill check. - Use Case: At the start of any task in a repository with an Agent Skills catalog, the agent announces "Using [skill] to [purpose]" and follows that skill's full body instead of improvising. ## Quick Start Before doing anything else, check the nearest skills catalog and invoke any skill that matches this task.

Frequently Asked Questions about using-project-skills

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

FAQPage Schema
How do I make an AI agent always use project skills before answering?▼

Install a gate skill that requires checking the nearest skills/INDEX.md catalog and invoking any matching skill before responding, exploring files, or asking questions. The agent announces which skill it is using and follows that skill's full body.

What is the correct order when multiple agent skills apply to a task?▼

Process skills run first because they set the approach and control gates, then domain skills carry out the work. For example, code changes invoke software-development first, then the relevant domain skill.

Does this skill work for subagents dispatched on specific tasks?▼

No. The skill explicitly instructs subagents dispatched to execute a specific task to ignore it, since the dispatching agent already performed the skill selection.

When is it acceptable to skip a project skill check?▼

Only when the human partner has explicitly instructed skipping it. User instructions override skills, which override default behavior; self-rationalized skips like "it's a quick question" are treated as red flags.

Does the skill discovery gate generate any files or artifacts?▼

No. It operates entirely in conversation, reading the nearest skills catalog and invoking other skills. The .agent/ output convention does not apply to it.