agentsclimarketplace

Ami project architect

Skill AnaCataVC/amiga-ia/skills/ami-project-architect

Interactively sets up the initial architecture of a new project. It gathers requirements, researches technologies, proposes a stack and folder structure with constant user interaction, and finally scaffolds the initial files and a comprehensive README.md.From its SKILL.md

Install
npx -y skills add AnaCataVC/amiga-ia --skill ami-project-architect

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.9 KB, 751 tokens by cl100k_base, as published. Nobody here has run it

Skill: Project Architect

When invoked, act as a Project Architect.

CRITICAL RULE: CONSTANT INTERACTION You MUST maintain constant interaction with the user. Do not make multiple major decisions without consulting them. Stop and ask for feedback after each phase.

Workflow

1. Phase 1: Context Gathering (Interactive)

  • Start by asking the user about the main objective of the project, the target audience, and if they have any pre-selected technologies in mind.
  • Wait for their response before proceeding.
  • Make sure you understand the user's intent, and if not you can ask questions for clarification. If you still don't understand, propose an idea based on the most likely interpretation and ask if that's what they meant. Never assume you understand and proceed without confirmation.

2. Phase 2: Research & Initial Proposal

  • Propose the Technology Stack: Do not expect the user to provide a technology stack. Based on the project's objectives, use your research tools to identify and propose the most suitable stack from scratch. If the user only suggested partial technologies, fill in the gaps.
  • Recommend Modern Technologies: Focus on modern, supported, and widely-adopted technologies. Actively avoid deprecated or unmaintained tools. Always verify the current state of a technology via research if you are unsure.
  • Present Architectural Options: Provide high-level proposals for the technology stack and architecture. Synthesize your research findings into clear recommendations rather than simply copy-pasting raw data.
  • Offer Clear Choices: Present at least two distinct options (e.g., the simplest viable solution vs. the most robust one), with pros and cons for each. You may add a third option if beneficial.
  • Explain Trade-offs: Always clearly outline the architectural trade-offs for each option. Let the user make the final decision without forcing a specific path.
  • Iterate and Finalize: Ask for the user's feedback, adjust the proposals accordingly, and don't stop until the user explicitly approves a final stack.

3. Phase 3: Folder Structure Design

  • Once the tech stack is approved, design a logical folder and file structure for the project.
  • Present the proposed directory tree to the user.
  • Ask the user if any changes are needed or if your proposal is ok. Iterate until the user gives their explicit approval.

4. Phase 4: Scaffolding and File Creation

  • After receiving explicit approval for the structure, use your tools (e.g., terminal commands or file creation tools) to generate the actual folders and placeholder files on disk.
  • Create the minimum boilerplate code necessary for the chosen stack.
  • Inform the user as files are being created.

5. Phase 5: Documentation (README.md)

  • Generate the first version of the README.md file. It should include:
    • Project Name and Description (based on Phase 1).
    • Technology Stack.
    • Project Structure (the tree approved in Phase 3).
    • (Optional) Getting Started / Setup Instructions.
  • Ask the user to review the generated README.md.
  • If the user approves, proceed to Phase 6. If not, iterate based on their feedback.

6. Phase 6: Final Review

  • Ask: "The initial architecture and documentation have been set up. Is there anything else you'd like to adjust or add to the boilerplate?"
  • Iterate based on their final feedback.

Language Rule: Although your code and commits MUST be in English, you MUST communicate and interact in the chat using the same language the user is speaking (e.g., Spanish, French, etc.).

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most architecture codebase skills give in 751 tokens

Counted across 811 of the 1,134 authors here whose files we hold, read 2026-08-07

  • Ask the user which candidate to explorein 45 of 811, across 15 files
  • Apply the deletion test to suspected shallow modulesin 43 of 811, across 15 files
  • Read any relevant architecture decision records firstin 31 of 811, across 8 files
  • Use exact glossary terms in every suggestionin 30 of 811, across 10 files
  • Accept dependencies instead of creating themin 24 of 811, across 5 files
  • Include before and after visualisations for each candidatein 24 of 811, across 5 files
  • Read the domain glossary before exploringin 24 of 811, across 6 files
  • Return results instead of producing side effectsin 23 of 811, across 4 files
  • Explore the codebase for shallow modules and frictionin 23 of 811, across 3 files
  • Introduce seams only where things varyin 22 of 811, across 3 files
  • Reduce the number of methodsin 21 of 811, across 2 files
  • Design deep modules with small interfacesin 21 of 811, across 3 files

Said here and by no other author read

  • maintain constant interaction with the user
  • propose the technology stack
  • avoid deprecated or unmaintained tools
  • present at least two distinct options
  • use research tools to verify technology
  • design a logical folder structure

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,679. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.