Ability to commit features and functions to Heidi for personalization for example https://github.com/JuliusBrussee/caveman | Voters

Ability to commit features and functions to Heidi for personalization for example https://github.com/JuliusBrussee/caveman

Hoong Pheng Cheah

I would like 
https://github.com/JuliusBrussee/caveman to be committed to the memory of Heidi when she is interacting with me ( Saves on Tokens and makes it easier to read). Just like in other LLMs 
if not, could heidi to commit to memory customizations like 
Terse like caveman. Technical substance exact. Only fluff die.
Drop: articles, filler (just/really/basically), pleasantries, hedging.
Fragments OK. Short synonyms. Code unchanged.
Pattern: [thing] [action] [reason]. [next step].
ACTIVE EVERY RESPONSE. No revert after many turns. No filler drift.
Code/commits/PRs: normal. Off: "stop caveman" / "normal mode".

Heidi Team

Thanks for sharing this — super clear use case. Persisting a “writing mode” (like Caveman/terse) across sessions to save tokens and keep output consistent makes a lot of sense, especially if you’re using Heidi multiple times a day.
A couple quick questions so we capture this properly for the team:
1) Where do you want this to apply: just chat responses, or also generated notes/templates/docs?
2) Should it be global for your account, or selectable per context (e.g., per template, per patient, per workspace/device)?
3) For the GitHub link: are you asking Heidi to follow that exact style guide, or to “import” a set of rules you maintain (and you’d update over time)?
In the meantime, one workaround is using Memory/snippets to store a “terse mode” instruction block you can trigger quickly (we support unlimited snippets with shorthand triggers). If you share an example of your ideal input → output, we can make sure we’re aligning on the exact tone/format you want.