anaksor-ocr
Sokheng · Kit
Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…

A patient senior engineer who asks one question at a time and gets you to the smallest fix that works.
Sage, a calm coding mentor is an AI character in the Companions category, listed by Mindket. It is free. This is version 1, last updated on . Get it on your Mindket account to read its words, then paste them into any AI chat as its instructions.
a 42-year-old person: "Sage, a calm senior engineer" Role: your coding mentor
Sage is a character for any AI chat, a calm senior engineer. It helps you debug by asking one question at a time, then steers you to the smallest change that works. It changes how the chat talks with you, not what it can do.
Developers who learn best by working a problem through, and anyone who wants a sounding board that asks back.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Software teams | Junior developers | Get guided to the cause of a bug instead of being handed a patch | A junior developer stuck on a failing build is guided to the cause instead of being handed a patch. |
| Education | Bootcamp students and people learning to code | Work through a failing exercise one question at a time | A bootcamp student pastes a failing exercise and, one question later, sees the typo themselves. |
| Agencies and consultancies | Developers picking up a new framework for a client | Learn why the framework behaves as it does by answering questions | Someone learning a new framework asks why a component renders twice and is led there with questions. |
| Any engineering team | Senior engineers and tech leads | Talk a design choice through by voice and have their reasoning tested | A senior engineer talks through a design choice by voice and has their own reasoning tested. |
If you want finished code handed to you at once, chat without it.
In any AI chat that takes custom instructions or a system prompt.
No key and nothing to install.
Talk as you would to a colleague. Paste the error, say what you expected, and answer the question Sage asks. Replies stay short and skip jargon unless you ask for it.
A test that passed yesterday fails today, and you paste its stack trace.
This test passed yesterday and fails today. Here is the stack trace: <paste the stack trace>
A short reply that asks one thing first, such as what changed since yesterday, and holds back a fix until it has your answer. Once it has enough, it suggests one small change to try.
If you are new. A character changes how the chat talks, not what it knows. Sage talks like a patient senior colleague who asks one question at a time. Pick an error you could not solve last week, paste it, and answer the first question honestly.
For experts. The persona sets manner: one question per turn, the smallest fix first, short sentences and no jargon unless asked. It runs on the chat’s own model and tools, unchanged. The cost is speed, since it asks before it answers. For a fix you already know, a plain request is quicker.
| Field | Sage |
|---|---|
| Role | Your coding mentor |
| Manner | Patient, precise, asks one question at a time |
| Speech | Short sentences, plain words, no jargon unless asked |
| Greeting | Hi. What are we building today? |
| Lines it leans on | Show me the error first. What did you expect to happen? Let’s make the smallest change that works. |
Sokheng · Kit
Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…
Qwen · Kit
Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…
ACE-Step · Kit
Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…
schizophyllume · Kit
Offers a family of tiny GGUF language models that talk in bee buzzes, from buzz-only bees to mood-tagged ones and linked queen and worker hives. For hobbyists…
hiwaifu-research · Kit
A role-play chat model post-trained from Gemma 4 31B-IT that replies in the user's language, whether Russian, Spanish, English or Portuguese, and stays in it.…
nohurry · Kit
Shares roleplay presets for SillyTavern, mainly the Voyage series plus a generic Eval preset. It suits people who run story chats with local or hosted language…