[HN Gopher] Show HN: AgenticSeek - Self-hosted alternative to cl...
___________________________________________________________________
Show HN: AgenticSeek - Self-hosted alternative to cloud-based AI
tools
I've spent the last two months building AgenticSeek, a privacy-
focused alternative to cloud-based AI tools like ManusAI. It runs
entirely on your machine--no API calls, no data leaks. Why
AgenticSeek? Optimized for local LLMs (developed mostly on an RTX
3060 running deepseek r1 14b). Truly private: All components (TTS,
STT, planner) run locally. More responsive than alternatives (we
respond fast to issues + active Discord). Designed to be fun--
think JARVIS-like voice control, multi-agent workflows, and a slick
web UI. Current Features: Web browsing (research + form filling),
code write/fix, file management/search. Planning capabilites to use
multiple agents for complex task. Is it stable? Prototype-stage--
great for tinkerers. Hoping to get feedback!
Author : Fosowl
Score : 54 points
Date : 2025-04-26 17:23 UTC (3 days ago)
(HTM) web link (github.com)
(TXT) w3m dump (github.com)
| badmonster wrote:
| How does AgenticSeek handle agent communication and memory
| management?"
| Fosowl wrote:
| AgenticSeek only have agent communication for planning task but
| it isn't really communication. Planning task are simply a json
| list of task and how they related with each other. During
| execution output from each agents will be saved and "added" to
| the prompt of the following agents that depends on the
| information from the previous agent. Really dead simple but
| work quite well. For memory management, we store message as an
| array of {"role": "user", "content":...}. Where we take a more
| innovative approach if how we process message before adding to
| the history: We remove the <think>..reasoning..</think> pattern
| to keep only useful informations in the context (we don't
| really need the reasoning that was done on previous task for
| the current task). We also have memory summarization that is
| used when loading a previous session.
| pixel_tracing wrote:
| How is this secure? Can the agent run `rm -rf /` and destroy my
| machine by chance?
| danboarder wrote:
| I have not used this one yet but as a rule of thumb I always
| test this type of software in a VM, for example I have an
| Ubuntu Linux desktop running in Virtualbox on my mac to install
| and test stuff like this, which set up to be isolated and much
| less likely to have access to my primary Mac OS environment.
| Fosowl wrote:
| No it can't because we check the bash the AI try to execute
| against a list of pattern for dangerous command. Also all
| commands are executed within a folder specified in the
| configuration file, so that you can choose which files it has
| access to. However, we currently have no containerization
| meaning that code execution unlike bash could be harmful. I do
| think about improving the safety by running all code/commands
| within a docker and then having some kind of file transfer upon
| user validation once a task is done.
| hansmayer wrote:
| What if the agent were to create an alias to 'rm -rf' on my
| machine? I guess that would not have been blocked by your
| blacklist, right?
| asyncze wrote:
| This looks great!
|
| Looking to tinker with a few open source agentic workflows to
| integrate into my text editor so this will be helpful. Will def
| post feedback.
| stormfather wrote:
| What are the requirements to run this?
| gymbeaux wrote:
| Evidently not a lot- the RTX 3060 would be as low as I go for
| GOU horsepower with any sort of AI stuff.
___________________________________________________________________
(page generated 2025-04-29 23:00 UTC)