Оффлайн
- Регистрация
- 1 Мар 2015
- Сообщения
- 1,481
- Баллы
- 155
You know that moment when you're 14 files deep in a codebase, three coffees in, and wondering if the grep gods are listening? I decided enough was enough — it was time to build a coding buddy that actually answers back. Enter ApolloAgent, my latest side project, now open-sourced and live on GitHub.
?
? What Is ApolloAgent?
ApolloAgent is a terminal-based AI code assistant that uses under the hood with:
...and just enough sarcasm to replace my inner monologue.
? Why I Built It (Spoiler: It’s a PoC for Learning)
This project is a proof of concept (PoC) I built mainly for studying how to implement an AI agent from scratch — exploring how to handle prompt engineering, tool use, function calls, memory, and decision-making logic.
It’s been my playground for understanding how multi-tool LLM assistants work under the hood. The goal wasn’t to ship a product, but to get my hands dirty, experiment, and learn.
That said... it’s functional, fun, and honestly pretty useful already ?
? Inside chat_operations.py
If ApolloAgent had a personality, this file would be its witty, well-organized prefrontal cortex.
is responsible for:
I even added an optional "thinking" delay. Because sometimes, you want your assistant to seem like it's pondering quantum mechanics when it’s really just parsing your typo.
?️ What's Next?
If you're curious, bored, or just really into Python metaprogramming, check it out and contribute. ApolloAgent is my humble attempt at turning dev tooling into a conversation — one that actually listens.
Got feedback or feature requests?
File an issue, open a PR, or just tell me it’s cool — I’m not above fishing for compliments.
?
? What Is ApolloAgent?
ApolloAgent is a terminal-based AI code assistant that uses under the hood with:
- ? Function calling
- ? Tool chaining
- ? Chat memory
File editing- ? Grep-style codebase searching
- ?️ A plugin-like tool registry
...and just enough sarcasm to replace my inner monologue.
? Why I Built It (Spoiler: It’s a PoC for Learning)
This project is a proof of concept (PoC) I built mainly for studying how to implement an AI agent from scratch — exploring how to handle prompt engineering, tool use, function calls, memory, and decision-making logic.
It’s been my playground for understanding how multi-tool LLM assistants work under the hood. The goal wasn’t to ship a product, but to get my hands dirty, experiment, and learn.
That said... it’s functional, fun, and honestly pretty useful already ?
? Inside chat_operations.py
If ApolloAgent had a personality, this file would be its witty, well-organized prefrontal cortex.
is responsible for:
- Structuring prompts (system + user)
- Deciding when to call tools vs. just answer
- Responding with function call results
- Falling back gracefully with "Sorry, I didn’t understand that" when all else fails
I even added an optional "thinking" delay. Because sometimes, you want your assistant to seem like it's pondering quantum mechanics when it’s really just parsing your typo.
?️ What's Next?
- ?️ Add voice input (so I can code while making espresso)
- ? Docker support for plug-and-play setup
- ? Sarcastic mode (already halfway there)
If you're curious, bored, or just really into Python metaprogramming, check it out and contribute. ApolloAgent is my humble attempt at turning dev tooling into a conversation — one that actually listens.
Got feedback or feature requests?
File an issue, open a PR, or just tell me it’s cool — I’m not above fishing for compliments.