Conversation
| AI agents on Triton | ||
| ------------------- | ||
|
|
||
| If you use AI coding agents (e.g. Claude Code, OpenAI Codex) on Triton or any other | ||
| computer connected to our infrastructure, please read our policy on acceptable use, | ||
| risks, and required cooperation: | ||
|
|
||
| :doc:`AI Agents on HPC <AI_agents_on_HPC>` |
There was a problem hiding this comment.
Is there a more sustianable way to have this? If we have a whole section here, there could as well be a section on not making too many small files, not making too short jobs, not opening /scratch in vscode, etc. Having a way to highlight important things at account creation seems like a good idea, but how to avoid it having be a parallel doc index?
I can think of either a list of important things, or a generic point "please scan docs for pages that are relevant to your use". We could also have a new box somewhere more prominent (triton docs page? sidebar?) that says the currently relevant stuff that is new and non-obvious. I had hoped that the Mastodon box could be this but we see how that went...
There was a problem hiding this comment.
I agree that maybe the link from usagepolicy is not the best since in the end this is not a policy, more about good practices. Should we move it under the list https://scicomp.aalto.fi/triton/#detailed-instructions ?
|
todo: add to toctree. I suggest under /triton/usage/ Suggested rename: /triton/usage/ai-agents.rst to match naming convention of other pages. It will then be included in the toctree automatically. |
2 participants
No description provided.