skills.rest
dvy1987
dvy1987COMMUNITY
Explore all skills by this author →
WebsiteGitHub

project-setup

Generate tailored AGENTS.md files by interviewing users and scanning repository artifacts.

3|Updated Apr 5, 2026
Search Tags:#onboarding#agents#orchestration#interview#agents.md#project-setup#bootstrap
One-click install
npx skills add https://github.com/dvy1987/agent-loom --skill project-setup-dvy1987
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: project-setup
Source: https://github.com/dvy1987/agent-loom/tree/main/.agents/skills/project-setup
Command: npx skills add https://github.com/dvy1987/agent-loom --skill project-setup-dvy1987

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Many projects lack a concise, agent-ready configuration that maps tasks, boundaries, and orchestration; this Skill eliminates uncertainty by producing a tailored AGENTS.md that orients agents to the project's goals, gaps, and conventions. It prevents generic or unsafe agent behaviour by forcing a brief interview, importing existing project context, and enforcing guardrails like a 150-line limit.

Core Features & Use Cases

  • Context-aware AGENTS.md generation: Interviews the user, scans repo artifacts (PRDs, package manifests, README), and synthesizes a single, project-specific AGENTS.md.
  • Orchestration Map & Boundaries: Produces a phase-based orchestration map, explicit Allowed/Ask First/Never boundaries, and key commands to make agents composable and safe.
  • Update Mode & Auditability: Supports UPDATE_ONLY mode for scoped edits, shows diffs before saving, and appends entries to a skill outputs log for traceability.
  • Use case: Onboard agents into a new repo by running this Skill to create a concise AGENTS.md that configures agent autonomy, test/build commands, and role-specific boundaries.

Quick Start

Ask the assistant to interview you about your role and project, then "generate an AGENTS.md for this project" to create and save a tailored AGENTS.md.

Frequently Asked Questions about project-setup

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

FAQPage Schema
How do I generate an AGENTS.md file for my project repository?▼

To generate an AGENTS.md, the Skill interviews you about your role and project, scans repo artifacts like PRDs and manifests, and synthesizes a tailored configuration file mapping agent boundaries and orchestration.

What is the best way to onboard AI agents into a new codebase?▼

Onboarding agents requires a concise AGENTS.md that orients them to project goals and conventions; this Skill forces a brief interview and scans your repository to produce safe, composable agent behavior guardrails.

Can I update an existing AGENTS.md without regenerating the entire file?▼

Yes, you can use UPDATE_ONLY mode to perform scoped edits on an existing AGENTS.md, showing diffs before saving and appending entries to a skill outputs log for traceability.

Are there line limits or size constraints for AGENTS.md files?▼

AGENTS.md files are strictly capped at a 150-line limit to prevent generic or unsafe agent behavior by forcing concise, project-specific orchestration maps and boundaries.

Does project-setup work with existing PRDs and package manifests?▼

Yes, the Skill performs silent repository scans to import existing project context from PRDs, package manifests, and README files to synthesize an accurate AGENTS.md configuration.

When should I regenerate an AGENTS.md configuration file?▼

You should regenerate AGENTS.md during project bootstrap, agent onboarding, repository initialization, or when PRDs, tech stacks, or team composition change and require updated agent behavior.

Related Skills

Explore More Software Engineering Skills→
kkestellkkestell
1

kinit

Generate an AGENTS.md file documenting project components, commands, and conventions.

Community
Intermediate
dyxushuaidyxushuai
2

agent-init

Create and maintain repo-wide AGENTS.md files for coding agents.

Community
Intermediate
AesopScottAesopScott
1

build-agent

Automate the MAPS Build phase from design artifacts into runnable agent loops.

Community
Advanced
aspenkitaspenkit
95

agent-customization

Inject project-specific context from AGENTS.md into installed agent templates.

Official
Advanced