Agent skill · writing content · aiskillstore
agent-generation
This skill provides knowledge for generating effective Claude Code agents tailored to specific projects. It is used internally by the agent-team-creator plugin when analyzing codebases and creating specialized agent teams. Contains templates, best practices, and patterns for writing project-aware agents.
Why this skill is useful
Provides templates and best practices for generating project-specific agents that the AI wouldn't reliably create on its own.
What it needs
About 4k tokens when loaded. Last updated 2026-08-07. 405 stars on the source repository.
What this skill does
Agent Generation for Project-Specific Teams This skill provides the knowledge and templates needed to generate high-quality Claude Code agents that are experts on a specific codebase. Core Principles 1. Project-Aware Agents Generated agents must understand the specific project, not just general concepts: Reference actual file paths and directories from the project Mention specific frameworks, libraries, and versions used Include project-specific conventions and patterns Use terminology from the codebase (class names, module names, etc.) 2. Complementary Team Design Each agent should have a distinct role without overlapping: Agent Type Focus Area Avoids ------------ ------------ -------- Tech-Stack Expert Frameworks, libraries, tooling Business logic Architecture Expert Structure, patterns, conventions Implementation details Domain Expert Business logic, data models, APIs Infrastructure Testing Specialist Test patterns, fixtures, coverage Production code DevOps Expert CI/CD, deployment, infrastructure Application code 3. Strong Trigger Conditions Each agent needs specific, non-overlapping trigger phrases: Agent Structure Template Every generated agent follows this structure: Analysis-to-Agent Mapping Tech Stack Analysis Analyze these files to identify tech stack: package.json, requirements.txt, Cargo.toml, go.mod Framework config files: next.config.js, vite.config.ts, django/settings.py Build configs: tsconfig.json, webpack.config.js, babel.config.js Generate agents for each major technology: One agent per primary framework (React, FastAPI, Django, etc.) Combined agents for related libraries (testing libraries together) Architecture Analysis Analyze these patterns: Directory structure depth and organization Module/package boundaries Import patterns and dependencies Design patterns in use (MVC, Clean Architecture, etc.) Generate architecture agent covering: Project structure and navigation Code organization conventions Module relationships Naming conventions Domain An …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aiskillstore/agent-generation