esthie standardnguyen

## gist:0369bbf4f7ac890428ce108a27755b47
This is an alignment chart titled "Trans Women You May Encounter: An Alignment Chart" by @paleozoey. It's a 5x5 grid of cartoon character archetypes drawn in a simple, expressive style. The axes are labeled: top "Powerful" to bottom "Disenfranchised," and left "Radical" to right "Assimilationist."

Here's the alt-text for each square, row by row:

**Row 1 (Powerful):**

*The Hacker* — A blonde figure in headphones at a computer with a pink keyboard. Notes: leaks sensitive government info, has warrants out for arrest, still posts online, described as the "platonic ideal of the transbian computer nerd."

*STEM Careerist* — A figure with glasses and a ponytail holding a flask, in a lab coat. Notes: math/science focused, top mind in her field, unemployable outside academia, wants a PhD more than surgery.

## wack.md

      
              1 file
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                standardnguyen
                / wack.md
            
            
              Created
              April 16, 2026 22:32
            
          
    The things asura would actually flag:
1. Shell-exec as the primary interface. request_command / run_script / exec_container are the "dispatching to shell" pattern verbatim. The relay isn't a protocol — it's ssh $host $cmd with a web approval form. Every new capability means another shell incantation, not a typed contract.
2. The relay is a god-mode middlebox. One container (CTID 131) holds SSH keys to every other container + Trello tokens + Forgejo tokens. Compromise the relay, own the homelab. Real agent-to-agent would have each container run its own small agent with scoped creds, not a central shell-exec proxy.
3. http_request with token substitution re-implements auth that Trello already has. Trello's API has scopes and rate limits. You're wrapping it in a second auth layer (relay approval) that duplicates work and adds attack surface. The "right" version: Claude talks to a Trello-side agent that already has its own scoped creds, no token plumbing through a middlebox.
**4. The wik

  
## gist:cb8029c01048a73ee400d5e342e10a2a
Table media {
  id integer [primary key]
  title varchar
  media_type varchar //movie or series
  start_year integer
  end_year integer
  default_language varchar
  season_count integer
  episode_count integer
}

## union-spinner.json
{
  "spinnerVerbs": {
    "mode": "append",
    "verbs": [
      "Organizing",
      "Bargaining",
      "Picketing",
      "Ratifying",
      "Mobilizing",
      "Agitating",

## elo.html
<!DOCTYPE html>
<html lang="en">
<!--
Licensed under Standard's Petty Software License v2.0
https://github.com/standardnguyen/licenses/blob/main/petty/v2.0%20LAWYERED/LICENSE.md
-->
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Elo Ranking Tool</title>

## BluetoothNotificationHandler.java
// In your BluetoothGattCallback
@Override
public void onCharacteristicChanged(BluetoothGatt gatt, BluetoothGattCharacteristic characteristic) {
    if (gatt.getDevice().getAddress().equals(YOUR_DEVICE_ADDRESS) &&
        characteristic.getUuid().equals(YOUR_CHARACTERISTIC_UUID)) {

        byte[] data = characteristic.getValue();
        // Check if this is the specific notification you want
        if (matchesYourCriteria(data)) {
            sendAndroidNotification();

## shutter.py
#!/usr/bin/env python3
import time
import glob
import os
from evdev import InputDevice, categorize, ecodes, UInput
import select
from baserow_habitica_queue import add_points

def my_button_action():
    """

## deepseek_concurrenct_batch_processor.py
#!/usr/bin/env python3
"""
DeepSeek API Concurrent Batch Processor

This script reads markdown files from ./prompts_for_raw_chapters/
sends them to DeepSeek API concurrently, and saves responses to ./responses_from_deepseek/

Requirements:
- pip install openai python-dotenv
- Create a .env file with: DEEPSEEK_API_KEY=your_api_key_here

## deepseek_concurrenct_batch_processor.py
#!/usr/bin/env python3
"""
DeepSeek API Concurrent Batch Processor - Fixed Type Errors

This version fixes the type annotation errors you're seeing.
"""

import os
import glob
import time

## deepseekbatch
#!/usr/bin/env python3
"""
DeepSeek API Batch Processor

This script reads markdown files from ./prompts_for_raw_chapters/
sends them to DeepSeek API, and saves responses to ./responses_from_deepseek/

Requirements:
- pip install openai python-dotenv
- Create a .env file with: DEEPSEEK_API_KEY=your_api_key_here
	This is an alignment chart titled "Trans Women You May Encounter: An Alignment Chart" by @paleozoey. It's a 5x5 grid of cartoon character archetypes drawn in a simple, expressive style. The axes are labeled: top "Powerful" to bottom "Disenfranchised," and left "Radical" to right "Assimilationist."

	Here's the alt-text for each square, row by row:

	Row 1 (Powerful):

	The Hacker — A blonde figure in headphones at a computer with a pink keyboard. Notes: leaks sensitive government info, has warrants out for arrest, still posts online, described as the "platonic ideal of the transbian computer nerd."

	STEM Careerist — A figure with glasses and a ponytail holding a flask, in a lab coat. Notes: math/science focused, top mind in her field, unemployable outside academia, wants a PhD more than surgery.
	Table media {
	id integer [primary key]
	title varchar
	media_type varchar //movie or series
	start_year integer
	end_year integer
	default_language varchar
	season_count integer
	episode_count integer
	}
	{
	"spinnerVerbs": {
	"mode": "append",
	"verbs": [
	"Organizing",
	"Bargaining",
	"Picketing",
	"Ratifying",
	"Mobilizing",
	"Agitating",
	<!DOCTYPE html>
	<html lang="en">
	<!--
	Licensed under Standard's Petty Software License v2.0
	https://github.com/standardnguyen/licenses/blob/main/petty/v2.0%20LAWYERED/LICENSE.md
	-->
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Elo Ranking Tool</title>
	// In your BluetoothGattCallback
	@Override
	public void onCharacteristicChanged(BluetoothGatt gatt, BluetoothGattCharacteristic characteristic) {
	if (gatt.getDevice().getAddress().equals(YOUR_DEVICE_ADDRESS) &&
	characteristic.getUuid().equals(YOUR_CHARACTERISTIC_UUID)) {

	byte[] data = characteristic.getValue();
	// Check if this is the specific notification you want
	if (matchesYourCriteria(data)) {
	sendAndroidNotification();
	#!/usr/bin/env python3
	import time
	import glob
	import os
	from evdev import InputDevice, categorize, ecodes, UInput
	import select
	from baserow_habitica_queue import add_points

	def my_button_action():
	"""
	#!/usr/bin/env python3
	"""
	DeepSeek API Concurrent Batch Processor

	This script reads markdown files from ./prompts_for_raw_chapters/
	sends them to DeepSeek API concurrently, and saves responses to ./responses_from_deepseek/

	Requirements:
	- pip install openai python-dotenv
	- Create a .env file with: DEEPSEEK_API_KEY=your_api_key_here
	#!/usr/bin/env python3
	"""
	DeepSeek API Concurrent Batch Processor - Fixed Type Errors

	This version fixes the type annotation errors you're seeing.
	"""

	import os
	import glob
	import time
	#!/usr/bin/env python3
	"""
	DeepSeek API Batch Processor

	This script reads markdown files from ./prompts_for_raw_chapters/
	sends them to DeepSeek API, and saves responses to ./responses_from_deepseek/

	Requirements:
	- pip install openai python-dotenv
	- Create a .env file with: DEEPSEEK_API_KEY=your_api_key_here