22 copy/paste-ready prompts for building your own AI agent system. Each prompt builds a functional system or implements a proven best practice you can hand to an AI coding assistant.
Replace placeholders like <your-workspace>, <your-messaging-platform>, and <your-model> with your own values.
Generalized versions of all root .md files used by OpenClaw. These files are loaded into the agent's system prompt on every request (except MEMORY.md which is conditional).
Copy these as starting points and customize for your own setup. Replace <placeholders> with your values.
| DEBUGGING: | |
| ——————————————————————————————————————— | |
| STOP. Before making any code changes, conduct a thorough root cause analysis. | |
| Using all available tools (codebase access, terminal, @web), investigate the issue systematically: | |
| 1. **Reproduce and observe** the exact error/behavior | |
| 2. **Trace the execution path** that leads to this issue | |
| 3. **Examine dependencies** and their versions if relevant | |
| 4. **Check assumptions** about how the code should work vs. how it actually works |
| Assorted AI resources/demo repos: | |
| - https://github.com/patchy631/ai-engineering-hub/tree/main/brand-monitoring | |
| AI Agents: | |
| - https://github.com/virattt/ai-hedge-fund/tree/main | |
| - https://github.com/HKUDS/AutoAgent/tree/main | |
| - https://github.com/coleam00/Archon/tree/main | |
| - https://github.com/OriNachum/autonomous-intelligence | |
| - deepresearch built with openai agents sdk and firecrawl ex. https://www.theunwindai.com/p/build-a-deep-research-agent-with-openai-agents-sdk-and-firecrawl | |
| - https://www.youtube.com/watch?v=U6LbW2IFUQw&sttick=0 |
| Research Prompt for Weekly Newsletter on Multimodal Models and Vision-Language Models | |
| Objective: | |
| Compile a detailed, structured overview of the latest developments in multimodal models and vision-language models (VLMs) from the past week. Include sections for Quick Take, Research Highlights, Tools & Techniques, Real-World Applications, Trends & Predictions, and Community Contributions. Ensure all content is fresh, excluding anything covered in prior newsletters, and tailor the output for a newsletter audience with concise, impactful summaries. | |
| General Instructions: | |
| Time Frame: Focus exclusively on developments from the past week, using a dynamic date range of [start date] to [end date], to be updated weekly. | |
| Uniqueness: Cross-reference with previous newsletters to avoid repetition, ensuring all content is new and relevant. | |
| Prioritization: Leverage OpenAI’s deep research capabilities to prioritize high-impact, technically substantive sources (e.g., detailed benchmarks, novel methodologies, or influenti |
| Google Gemini | |
| - https://x.com/kaiju_ya/status/1900363384870363423 |
| from covalent import CovalentClient | |
| import csv | |
| from datetime import datetime, timedelta | |
| # Constants | |
| API_KEY = 'ckey_1bc79adaaaf24ab59fd0d566f25' # Replace this with your actual API key | |
| CHAIN_ID = 'eth-mainnet' | |
| CONTRACT_ADDRESS = '0xd417144312dbf50465b1c641d016962017ef6240' # Covalent Token address | |
| TODAY = datetime.now() | |
| TWO_MONTHS_AGO = TODAY - timedelta(days=60) |
| #!/bin/bash | |
| # This script fixes helmcharts secrets that was deployed with API objects paths, that was deprecated. | |
| # More info here https://github.com/helm/helm/issues/7219 | |
| # Use it own risks! | |
| # After executing you need to redeploy | |
| NamespaceName=$1 | |
| echo "${NamespaceName}" |
| apiVersion: apps/v1 | |
| kind: Deployment | |
| metadata: | |
| name: cluster-autoscaler | |
| namespace: kube-system | |
| labels: | |
| app: cluster-autoscaler | |
| spec: | |
| replicas: 1 | |
| selector: |