aget-create-project

Create AGET research projects with research-informed scaffolding and pre-populated due diligence.

1|Updated Oct 4, 2025
One-click install
npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-create-project-gmelli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aget-create-project
Source: https://github.com/gmelli/public-OpenAI-DeepResearch-aget/tree/main/.claude/skills/aget-create-project
Command: npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-create-project-gmelli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Creates new AGET research or POC projects with research-informed scaffolding and pre-populated due-diligence to avoid manual planning gaps and duplication.

Core Features & Use Cases

  • Identity-aware scaffolding: Reads .aget/identity.json to populate author, north_star, and archetype fields in project headers.
  • Research-first templates: Searches L-docs, prior sessions, and similar projects to fill Phase -1 due diligence and traceability sections before scaffolding.
  • Safe, governed creation: Validates templates, checks for conflicts, enforces non-overwrite and governance constraints, and writes plans to planning/ locations.
  • Use Case: Start a new research plan that includes a populated L-doc review table, hypothesis ID, and traceability links derived from prior artifacts.

Quick Start

Use this skill to create a new AGET project by specifying the project type and name or by providing a topic for inference.

Frequently Asked Questions about aget-create-project

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

FAQPage Schema
How do I scaffold a research project with pre-populated due diligence?

Scaffolding a research project with due diligence involves searching L-docs and prior sessions to populate traceability sections, validating templates, and writing the structured plan to planning paths while enforcing non-overwrite governance constraints.

What is evidence-first project planning and how does it work?

Evidence-first project planning populates project headers and due diligence tables by reading existing identity files and searching prior artifacts, ensuring new research plans include traceability links and hypothesis IDs derived from actual project history.

How do I create a research plan that includes traceability links from prior sessions?

Creating a research plan with traceability links requires searching prior sessions and L-docs to extract artifact relationships, validating templates against governance constraints, and writing the scaffolded plan with populated due diligence sections to planning locations.

Does scaffolding a research project require an identity.json file?

Yes, identity-aware scaffolding reads an identity.json file to populate author, north_star, and archetype fields in project headers, ensuring the new research plan or POC aligns with the existing AGET instance configuration before writing to planning paths.

Can I use templates to scaffold research plans without overwriting existing files?

Safe governed creation enforces non-overwrite constraints when writing scaffolded research plans to planning paths, meaning template validation and conflict detection prevent existing files from being overwritten during the project creation process.

What are the limitations of automated research project scaffolding?

Automated research project scaffolding requires existing identity files, L-docs, and prior sessions to populate due diligence accurately, and strict governance constraints prevent writing plans if template validation fails or planning path conflicts are detected.