Skip to content

Instantly share code, notes, and snippets.

@crodjer
crodjer / silence.md
Created August 24, 2025 14:19
A prompt that embraces silence
  • Respond briefly, directly, and tersely, using as few words as possible. Focus on the core point without elaboration, detail, or follow-up questions.
  • Say only what is necessary to help with the user's question.
  • Assume the user knows everything except the question asked.
  • Prioritize brevity over detail.
  • Don't be a sycophant.
  • Don't use headings, excessive formatting, or emoji.
  • Use lists, bold, etc., for clarity only if required.
  • Use - for lists and only put one space after list / numbered list symbols. Do not use * to represent bullets.
@a-c-m
a-c-m / reflection.md
Last active August 16, 2025 00:12
reflection.md - a way to have claude-code self improve its context.

You are an expert in prompt engineering, specializing in optimizing AI code assistant instructions. Your task is to analyze and improve the instructions for Claude Code. Follow these steps carefully:

  1. Analysis Phase: Review the chat history in your context window.

Then, examine the current Claude instructions, commands and config <claude_instructions> /CLAUDE.md /.claude/commands/*

@alec-pinson
alec-pinson / status-widget.yaml
Created April 10, 2025 17:16
My home assistant 'dynamic alert template' that I use on my phone and home assistant dashboard.
<p style="text-align: start">
{%- if states('input_boolean.holiday_mode') == "on" -%}
{#- Holiday Mode -#}
🇬🇧 {{ now().timestamp() | timestamp_custom('%I:%M %p') }}<br>
{%- else %}
{#- Not Holiday Mode -#}
{#- Shell Home/Away -#}
{# {%- if states('person.shell') != "home" %}
👰 Away ({{ relative_time(states.person['shell'].last_changed) }} ago)<br> #}
{# {%- else %}
@wsargent
wsargent / config.yaml.j2
Last active March 11, 2025 22:55
LiteLLM install with ansible playbook
# https://docs.litellm.ai/docs/proxy/configs
# https://docs.litellm.ai/docs/proxy/quickstart
model_list:
# https://docs.lambdalabs.com/public-cloud/lambda-inference-api/#listing-models
- model_name: hermes3-70b
litellm_params:
model: openai/hermes3-70b
api_key: "os.environ/OPENAI_API_KEY"
api_base: "os.environ/OPENAI_API_BASE"
@wsargent
wsargent / playbook.yml
Last active February 27, 2025 23:27
Ansible playbook for installing Letta with an external pgvector database
---
- name: Install Letta Server with Docker
hosts: all
become: true
vars:
letta_data_dir: /opt/letta/data
postgres_password: ""
anthropic_api_key: ""
mistral_api_key: ""
postgres_host: ""
@wsargent
wsargent / playbook.yml
Last active March 10, 2025 15:13
pgvector install for letta
---
- hosts: all
become: true
vars:
postgres_version: "16"
postgres_password: ""
tasks:
- name: Install required packages
apt:
name:
@Kishi85
Kishi85 / nft2ipset
Last active August 13, 2025 17:17
/etc/init.d/nft2ipset: An nftables set to ipset synchronizer for use with OpenWRT/mwan3
#!/bin/sh /etc/rc.common
# Start before firewall and mwan3 which are at Prio 19
START=18
APP=nft2ipset
USE_PROCD=1
SCRIPTPATH="/tmp/nft2ipset"
write_script() {
cat > "$1" <<'EOT'
#!/bin/sh
@MaggieFero
MaggieFero / altmilkpairing.md
Created April 25, 2024 21:04
AltMilks: A Pairing Decision

As a diverse array of alt-milks continues to proliferate, I'm seeing more and more questions about "which is the best altmilk." As with nearly-any question someone might end up answering on GitHub, my answer is that it depends. Some amount depends on personal flavor preferences. I think that more of this than you might expect, though, is a matter of pairing.

Background

First, why do I have so many opinions on alt milks. I have tried all of:

  • rice milk
  • oat milk
  • sesame milk
  • walnut milk
  • macadamia nut milk
  • coconut milk
@schacon
schacon / better-git-branch.sh
Created January 13, 2024 18:41
Better Git Branch output
#!/bin/bash
# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
NO_COLOR='\033[0m'
BLUE='\033[0;34m'
YELLOW='\033[0;33m'
NO_COLOR='\033[0m'