Agent skills recommendations that are compatible with this template?

[Allena101]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question and didn't find it.
• I searched the documentation, with the integrated search.
• I already searched in Google "How to X" and didn't find any information.
• I already read and followed all the tutorials in the docs and didn't find an answer.

Example Code

None

Description

I can sese in the .claude folder that there is a fastapi and sqlmodel skill (if i understand it correctly). Which other skills how you guys found to be very usefull for the template to work very well with an AI assistant?

For example. A common skill recommendation is the frontend-design skill on skills.sh . This made me think about how compatible it would be with the frontend part stack that this template uses.

Operating System

Linux

Operating System Details

No response

Project Version

Current

Python Version

No response

Additional Context

No response

[wu736139669]

Two answers, because they solve different problems.

For the template stack itself: keep the built-in FastAPI/SQLModel skills and add a frontend skill only if you're changing the UI. frontend-design is a reasonable default, as you said. Those are the skills that know the stack.

The one general skill I'd add is not template-specific: Goal Instruction Writer (Codex). It makes the agent write down the outcome, acceptance evidence, scope, and stop conditions before a multi-step change, then keep a checkpoint between turns. That helps when one change spans backend, frontend, migrations, and tests, which is where agents tend to silently expand scope. It's instruction-only and does not replace the template's own skills or the frontend skill; skip it if your tasks are one-shot edits.

If you try it with this template, the useful check is whether the generated Goal names the actual acceptance evidence (for example, a passing backend test plus a real browser flow) rather than "looks done".

[Allena101]

Two answers, because they solve different problems.

For the template stack itself: keep the built-in FastAPI/SQLModel skills and add a frontend skill only if you're changing the UI. frontend-design is a reasonable default, as you said. Those are the skills that know the stack.

The one general skill I'd add is not template-specific: Goal Instruction Writer (Codex). It makes the agent write down the outcome, acceptance evidence, scope, and stop conditions before a multi-step change, then keep a checkpoint between turns. That helps when one change spans backend, frontend, migrations, and tests, which is where agents tend to silently expand scope. It's instruction-only and does not replace the template's own skills or the frontend skill; skip it if your tasks are one-shot edits.

If you try it with this template, the useful check is whether the generated Goal names the actual acceptance evidence (for example, a passing backend test plus a real browser flow) rather than "looks done".

Thanks for the reply and info Wu!

Another thing I wondered when it comes to implementing a code assistant with this template. Should I provide the documentation to this template repo? Or is it perhaps enough for the LLM to understand this without?

Maybe , if necessary, you could save some template instructions in an MD file and include it in the prompt i.e. @template_workflow or something