Phillip Ninan fourgates

See more of my writing here.

In this post, I'll start from scratch and build up to OpenClaw's architecture step by step, showing how you could have invented it yourself from first principles, using nothing but a messaging API, an LLM, and the desire to make AI actually useful outside the chat window.

End goal: understand how persistent AI assistants work, so you can build your own (or become an OpenClaw power user).

First, let's establish the problem

When you use ChatGPT or Claude in a browser, there are several limitations:

Delete unused or obsolete files when your changes make them irrelevant (refactors, feature removals, etc.), and revert files only when the change is yours or explicitly requested. If a git operation leaves you unsure about other agents' in-flight work, stop and coordinate instead of deleting.
Before attempting to delete a file to resolve a local type/lint failure, stop and ask the user. Other agents are often editing adjacent files; deleting their work to silence an error is never acceptable without explicit approval.
NEVER edit .env or any environment variable files—only the user may change them.
Coordinate with other agents before removing their in-progress edits—don't revert or delete work you didn't author unless everyone agrees.
Moving/renaming and restoring files is allowed.
ABSOLUTELY NEVER run destructive git operations (e.g., git reset --hard, rm, git checkout/git restore to an older commit) unless the user gives an explicit, written instruction in this conversation. Treat t

dspy.Adapter

::: dspy.Adapter handler: python options: members: - call - acall

format

graph TD
  A[Should I use an Angular module?] --> B{Is it a component?}
  B -->|Yes| X[Don't use an Angular module]
  B -->|No| C{Is it a directive?}
  C -->|Yes| X
  C -->|No| D{Is it a pipe?}
  D -->|Yes| X
  D -->|No| E{Is it a service?}
 E --&gt;|Yes| X

When it comes to databases and AWS VPC, best practice is to place your database in private subnet. By definition, private subnet in AWS is not reachable from the Internet because no Internet gateway is attached to private subnet. This is the way you protect your data. This kind of configuration is good for security but bad for data management.

How can you easily access and manage your secured data?

There are two basic ways to acees it.

Access postgres RDS from bastion host. There are following requirements for this.

	@Get('api/auth/sso/saml/login')
	@UseGuards(SamlAuthGuard)
	async samlLogin() {
	//this route is handled by passport-saml
	return;
	}

	import * as cdk from "@aws-cdk/core";
	import * as wafv2 from "@aws-cdk/aws-wafv2";

	// This extends the base cdk stack properties to include a tag name input.
	export interface StackProps extends cdk.StackProps {
	tag: string;
	applicationName?: string;
	}

	export class WAFStack extends cdk.Stack {

	#
	# This small example shows you how to access JS-based requests via Selenium
	# Like this, one can access raw data for scraping,
	# for example on many JS-intensive/React-based websites
	#

	from time import sleep

	from selenium import webdriver
	from selenium.webdriver import DesiredCapabilities

	import { Router } from "express";
	import jwt from "jsonwebtoken";
	import jwkToPem from "jwk-to-pem";
	import * as Axios from 'axios';

	interface PublicKeys {
	keys: PublicKey[];
	}
	interface PublicKey {
	alg: string;

	import cdk = require("@aws-cdk/core");
	import { Vpc, Port } from "@aws-cdk/aws-ec2";
	import {
	Cluster,
	ContainerImage,
	AwsLogDriver,
	FargatePlatformVersion,
	NetworkMode,
	CfnService,
	} from "@aws-cdk/aws-ecs";