Nostr notes by owner

Feel free to join this test channel: ...

2026-04-16T13:16:38Z

Feel free to join this test channel:
https://smp2.simplexonflux.com/c#2y8JODOYPFq6ICpfBels-mNOFneumoZ89SHhNiCluj8

#simplex #simplex-chat #simplexchat #privacy

White noise is more Nostr then Signal

2025-11-06T22:52:16Z

In reply to nevent1q…6jwp
_________________________

White noise is more Nostr then Signal

Do you use SimpleX aswell btw.?

2025-09-20T11:26:32Z

In reply to nevent1q…0tpg
_________________________

Do you use SimpleX aswell btw.?

@nprofile…9kek ping

2025-09-20T02:30:53Z

In reply to nevent1q…ny4x
_________________________

TrueDeMoN100 (nprofile…9kek) ping

Rewrote my bot to read values from a config file and the result ...

2025-09-18T19:20:20Z

Rewrote my bot to read values from a config file and the result has been great so far. Here is how it looks for me:
[Sampler]
temperature = 0.62
max_length = 14500
max_context_length = 32768
top_p = 0.90
min_p = 0.07
top_k = 25
rep_pen = 1.04
rep_pen_range = 330
rep_pen_slope = 0.6
enable_thinking = False
stop_sequence = <end_of_turn>\n<start_of_turn>user, <end_of_turn>\n<start_of_turn>model, <end_of_turn>
[Settings]
host = http://localhost:5001
display_name=UEAIB
full_name=unofficial experimental AI bot
description=I'm not responsible for alien invasions
welcome_message=This is a test chatbot running on my shitbox. I can see all of your queries.\nUse /reset command to start a new session. Session will automatically reset after 5 minutes without messages.\nDo NOT share private or security-sensitive information!\ntemporary contact: https://smp5.simplex.im/a#-m-PgMG-4tFTrzkwF3BFf5x-rDHbaW7oeFOWNrhLtFs
auto_accept_message=
command_prefix=/
image = data:image/jpg;base64,/9j/4AAQSkZJRgABAQAAAQABAA(long base64 image)
[Adapter]
system_start: <|im_start|>system\n
system_end: <|im_end|>\n
user_start: <|im_start|>user\n
user_end: <|im_end|>\n
assistant_start: <|im_start|>assistant\n
assistant_end: <|im_end|>\n

I'm too lazy to write anything else, just beleave me that this is exciting.

Link:
https://smp6.simplex.im/a#cSOiK--O58xQht8Nz5B0-qygNSl-l3IAVN6rJ6qTGeI

#simplex
#SimpleX
#bot
#messenger
#privacy
#nost
#introductions
Profile picture by Vyram Kraven (nprofile…r8wp)

I've made updates and improvements to the unofficial ...

2025-09-18T05:49:30Z

I've made updates and improvements to the unofficial experimental AI bot, so that now, fox ex., the /reset command is terminating the generation task, while running all commands asyncronously.

https://smp6.simplex.im/a#cSOiK--O58xQht8Nz5B0-qygNSl-l3IAVN6rJ6qTGeI

Why?

2025-09-16T00:00:00Z

In reply to nevent1q…nh5m
_________________________

Why?

Are you from Nostr or from SimpeX btw?

2025-09-15T23:28:57Z

In reply to nevent1q…ry38
_________________________

Are you from Nostr or from SimpeX btw?

Can be invited into groups btw, though be aware: It's uptime ...

2025-09-15T23:16:38Z

In reply to nevent1q…r98g
_________________________

Can be invited into groups btw, though be aware: It's uptime is not guaranteed.

Check out my @nprofile…nguq LLM bot using koboldcpp. It's ...

2025-09-15T23:15:24Z

Check out my SimpleX Chat (nprofile…nguq) LLM bot using koboldcpp. It's pretty self explanetory, here is the link:

https://smp6.simplex.im/a#cSOiK--O58xQht8Nz5B0-qygNSl-l3IAVN6rJ6qTGeI

Ps: I can see what you type into there ofc

#simplex
#SimpleX
#privacy
#technology
#messenger
#messengers
#datenschutz

@nprofile…576c

2025-09-15T20:48:59Z

zenobit (nprofile…576c)

stop it

2025-09-15T20:47:37Z

In reply to nevent1q…zw2u
_________________________

stop it

Es gibt eine enterprise version von Windows

2025-09-08T00:03:34Z

In reply to nevent1q…axas
_________________________

Es gibt eine enterprise version von Windows

Cool

2025-09-06T01:06:43Z

In reply to nevent1q…zay7
_________________________

Cool

Go ahead

2025-09-06T01:02:53Z

In reply to nevent1q…fsj8
_________________________

Go ahead

@nprofile…nguq

2025-09-06T00:48:11Z

In reply to nevent1q…4s24
_________________________

SimpleX Chat (nprofile…nguq)

Qubes/whonix vielleicht?

2025-09-06T00:46:41Z

In reply to nevent1q…67yn
_________________________

Qubes/whonix vielleicht?

Ich denke, das ist ein wenig zu weit gegriffen, aber es scheint ...

2025-09-06T00:46:04Z

In reply to nevent1q…vvq4
_________________________

Ich denke, das ist ein wenig zu weit gegriffen, aber es scheint real zu sein.

Cool, dass ihr auf Nostr seit :).

2025-09-06T00:38:33Z

In reply to nevent1q…zx6l
_________________________

Cool, dass ihr auf Nostr seit :).

As you're trying alternatives to Mastodon: What do you ...

2025-09-06T00:30:52Z

In reply to nevent1q…0zz4
_________________________

As you're trying alternatives to Mastodon: What do you expect?

Is the source code already publicly available on github?

2025-09-06T00:29:46Z

In reply to nevent1q…j6t0
_________________________

Is the source code already publicly available on github?

Mastodon is much more restrictive as a federated Network then ...

2025-09-06T00:19:02Z

In reply to nevent1q…83cy
_________________________

Mastodon is much more restrictive as a federated Network then Nostr. Be aware both aren't designed for privacy in particular, but for freedom of speech.

You probably can

2025-09-06T00:06:38Z

In reply to nevent1q…knz0
_________________________

You probably can

mischt

2025-09-06T00:04:20Z

In reply to nevent1q…tmhz
_________________________

mischt

Lol

2025-09-06T00:03:18Z

In reply to nevent1q…uru8
_________________________

Lol

It's the personal blog of Evgeny, who is the founder. Ehem, ...

2025-09-04T22:42:06Z

In reply to nevent1q…mx40
_________________________

It's the personal blog of Evgeny, who is the founder.
Ehem, OUR founder :D.

Many, but not all modifications can be done afterwards.

2025-09-04T21:52:56Z

In reply to nevent1q…rfq3
_________________________

Many, but not all modifications can be done afterwards.

Building yourself is the way, but good luck with all the stuff ...

2025-09-04T21:50:03Z

In reply to nevent1q…ej86
_________________________

Building yourself is the way, but good luck with all the stuff you need to compile.

oof

2025-09-04T19:24:06Z

In reply to nevent1q…r6c6
_________________________

oof

Hello I'm owner and I like to say that I have a lot of time ...

2025-09-04T10:54:43Z

Hello I'm owner and I like to say that I have a lot of time to waste even though I don't. I'm currently having a bit of fun with writing chatbots for SimpleX Chat (nprofile…nguq).
I'd really appreciate social interaction. I dont want fame, I just want to discuss.

#introductions
#SimpleX
#privacy

Updated the bot with this code to accept group invites ...

2025-09-03T19:37:23Z

Updated the bot with this code to accept group invites automatically:

@bot.event('receivedGroupInvitation')
async def on_contact_request(chat_info):
chat_id = str(chat_info["groupInfo"]["groupId"])
#print("CHAT_ID: "+chat_id)
await bot.client.api_join_group(chat_id)

#SimpleX #privacy #python #ai

@nprofile…ksdv Hast du eine meinung zu MLS? ...

2025-09-03T11:05:23Z

Kuketz-Blog 🛡 (nprofile…ksdv) Hast du eine meinung zu MLS? https://www.rfc-editor.org/rfc/rfc9420.html
https://www.poberezkin.com/posts/2025-08-12-mls-the-naked-king-of-end-to-end-encryption.html https://soatok.blog/2025/08/25/barking-up-the-ratchet-tree-mls-is-neither-royal-nor-nude/

Modified code to self-host the bot mentioned in my profile with ...

2025-09-03T10:05:26Z

Modified code to self-host the bot mentioned in my profile with koboldcpp:

import asyncio
from simpx import BotProfile, SimpleXBot
from simpx.download import SimpleXDaemon
from simpx.extension import ChatWrapper
from datetime import datetime, timedelta, timezone
import traceback
import os
import time
import logging

chat_histories: dict[str, dict] = {}

try:
from openai import OpenAI

aiclient = OpenAI(
base_url="http://localhost:5001/v1/";,
api_key="KOBOLDCPP",
)
except ImportError:
aiclient = None
print("ai client import error")

CHAT_MODEL = "koboldcpp/mlabonne_Qwen3-0.6B-abliterated-Q3_K_L"

# Example usage
if __name__ == "__main__":

daemon = SimpleXDaemon()
daemon.run(port_number=5225)

# Create a bot profile
profile = BotProfile(
display_name="SimpleX Chat AI bot",
full_name="",
description="",
welcome_message="""This is a test chatbot connected to [FluxAI](https://ai.runonflux.com/). You can use it for technical and other questions.

Use /reset command to start a new session. Session will automatically reset after 5 minutes without messages.

Do NOT share private or security-sensitive information!""",
auto_accept_message="",
command_prefix="/",
)

# Create the bot with the profile
bot = SimpleXBot(profile)

@bot.command(name="info", help="Shows bot information")
async def info_command(chat_info, profile):
"""Command that shows information about the bot's profile."""
await bot.send_message(
chat_info,
f"*Bot Information*\n"
f"Name: {profile.display_name}\n"
f"Description: {profile.description}\n"
f"Address: {profile.address}",
)

@bot.command(name="reset", help="Reset the conversation history for this chat")
async def reset_command(chat_info):
chat_type = chat_info.get("type")
if chat_type == "direct":
chat_id = chat_info["contact"]["contactId"]
elif chat_type == "group":
chat_id = chat_info["groupInfo"]["groupId"]
else:
return

chat_histories[chat_id]["messages"] = [] # wipe the list
await bot.send_message(chat_info, "Conversation history has been reset.")

if aiclient is not None:

@bot.on_message()
async def handle_plain_message(text: str, chat_info: dict, chat_item: dict):
"""
Demonstrate live messaging with an AI streaming response from OpenRouter.

Args:
chat_info: Chat information dictionary.
args: User's prompt.
"""

# Create a ChatWrapper instance using the provided chat_info and the bot's client.
chat = ChatWrapper(chat_info, bot.client)

chat_type = chat_info.get("type")
if chat_type == "direct":
chat_id = chat_info["contact"]["contactId"]
elif chat_type == "group":
chat_id = chat_info["groupInfo"]["groupId"]
else:
return

now = datetime.now(timezone.utc)

# Initialize history entry if it doesn't exist
if chat_id not in chat_histories:
chat_histories[chat_id] = {"messages": [], "last_seen": now}
else:
# Check if 5 minutes have passed since the last message
last_seen = chat_histories[chat_id]["last_seen"]
if now - last_seen >= timedelta(minutes=5):
# Clear the conversation history
chat_histories[chat_id]["messages"] = []

# Update last_seen timestamp
chat_histories[chat_id]["last_seen"] = now

# Add the new user message to the history
chat_histories[chat_id]["messages"].append(
{"role": "user", "content": text}
)

try:
# Start a live message session with an initial text.
initial_text = "Thinking..."
live_msg = await bot.send_message(chat, initial_text, live=True, ttl=60)
current_response = ""

# Create the chat completion request with streaming enabled.
response = aiclient.chat.completions.create(
model=CHAT_MODEL,
messages=chat_histories[chat_id]["messages"],
stream=True, # Enable streaming of the response.
)

# Buffer to accumulate tokens until a sentence is complete
buffer = ""
last_update = time.time()
UPDATE_INTERVAL = 2 # seconds

async def flush_buffer(force: bool = False):
"""Push complete sentences to the live message at UPDATE_INTERVAL-second cadence."""
nonlocal buffer, current_response, last_update
now = time.time()

if force or (
buffer.rstrip().endswith((".", "?", "!"))
and now - last_update >= UPDATE_INTERVAL
):
current_response += buffer
await live_msg.update_live(current_response.strip())
buffer = ""
last_update = now

# Process each streaming chunk.
for chunk in response:
# Extract text content from the current chunk.
# Assumes chunk structure similar to OpenAI's streaming response.
chunk_text = chunk.choices[0].delta.content or ""
if "<|eot_id|>" in chunk_text:
chunk_text = chunk_text.replace("<|eot_id|>", "")
if chunk_text: # keep any leftover text
buffer += chunk_text
break # no more data expected
buffer += chunk_text
await flush_buffer()

# Final flush for any remaining text
await flush_buffer(force=True)

chat_histories[chat_id]["messages"].append(
{"role": "assistant", "content": current_response}
)
# Finalize the live message.
await live_msg.finish_live()

except Exception as e:
# In case of error, finalize the live message.
traceback.print_exc()
await live_msg.finish_live()

# Start the bot
try:
asyncio.run(bot.start())
except KeyboardInterrupt:
logging.info("Exiting")

L'application povez checker du message continûment aussi.

2025-09-03T00:44:16Z

In reply to nevent1q…fhh5
_________________________

L'application povez checker du message continûment aussi.

SimpleX - Chat introduced an experimental LLM-bot which allows ...

2025-09-02T23:24:36Z

SimpleX - Chat introduced an experimental LLM-bot which allows users to interect with AI through chat. It is built upon the yet unstable python framework simpx-py. The bot is making use of the OpenAI python module to stream requests using live messages. Self hosting is currently possible with koboldcpp thanks to its built-in OpenAI API emulation. Be aware that the bot can read all of your messages.

https://github.com/shumvgolove/simpx-py/tree/fixes

https://github.com/simplex-chat/simplex-chat/blob/sh/llm-bot/apps/simplex-llm-bot/bot.py

https://github.com/LostRuins/koboldcpp

#SimpleX #privacy

Nostr event nevent1qqsxzfggrpf4u7kvly8afm9jm2rzc3e6zdlv7x575qd467xqhvz2q9szypzr9jeq4j5kgxjnply0s0ek46dd76utrkmg8ad7sw9gqcqq0qxuk570lkz

2025-09-02T23:10:11Z

Hello