Vince Fulco--Bighire.tools weisisheng

Final Review - Comprehensive PR Review & Testing

Step 0: Determine Review Pass

Before starting, check the git history to determine if this is a follow-up review:

git log --oneline -10 | grep -i "Co-Authored-By: Claude"

Building AI agents with the Claude Agent SDK

If you've used Claude Code, you've seen what an AI agent can actually do—read files, run commands, edit code, figure out the steps to accomplish a task.

And you know it doesn't just help you write code, it takes ownership of problems and works through them the way a thoughtful engineer would.

The Claude Agent SDK is the same engine, yours to point at whatever problem you want, so you can easily build agents of your own.

The Claude Agent SDK is how you build that same thing into your own applications.

🥫 The Ketchup Technique

Controlled Bursts | TDD | TCR | 100% Coverage | Emergent Design | Extreme Ownership

Core Loop

Red → Green → TCR → Refactor → TCR → Done
TCR: test && commit || revert
- Pass → commit → continue

Claude Code

Plan Mode

Shift + tab + tab
Use “opus” for planning and Sonnet for everything else, /model

Multi Agent Parallel Tasks

Agentic Coding MCPs

Overview

Powered by composio this MCP.json provides detailed information on Model Context Protocol (MCP) integration capabilities and enables seamless agent workflows by connecting to more than 80 servers.

It covers development, AI, data management, productivity, cloud storage, e-commerce, finance, communication, and design. Each server offers specialized tools, allowing agents to securely access, automate, and manage external services through a unified and modular system. This approach supports building dynamic, scalable, and intelligent workflows with minimal setup and maximum flexibility.

	Deno.bench("Deno, Bun, Node.js TypeScript Benchmarks", async (b) => {
	// Open a file that we will act upon.
	// using file = await Deno.open("a_big_data_file.txt");

	// Tell the benchmarking tool that this is the only section you want
	// to measure.
	b.start();

	// Now let's measure how long it takes to read all of the data from the file.
	// await new Response(file.readable).arrayBuffer();

	-- Query the API directly and flatten the nested JSON structure
	WITH raw_data AS (
	SELECT * FROM read_json_auto('https://public.api.bsky.app/xrpc/app.bsky.feed.getAuthorFeed?actor=did:plc:edglm4muiyzty2snc55ysuqx&limit=10')
	),
	unnested_feed AS (
	SELECT unnest(feed) as post_data FROM raw_data
	)
	SELECT
	-- Post basics
	post_data.post.uri as post_uri,

	import { getZodConstraint, parseWithZod } from "@conform-to/zod"
	import type { HTMLAttributes } from "astro/types"
	import { ZodAny, z, type ZodIssue } from "zod"

	export type Constraint = ReturnType<typeof getZodConstraint>[0]
	export type FieldAttributes = {
	name: string
	value: string
	type: NonNullable<HTMLAttributes<"input">["type"]>
	} & Partial<Record<Lowercase<keyof Constraint>, any>>

	"use client";

	import { useEffect } from 'react';

	// Note: Logout works with a direct link to NextAuth's unbranded /api/auth/signout
	// however signOut does not appear to work consistently (e.g. doesn't clear session) and may cause redirect loops

	async function fetchCsrfToken() {
	const response = await fetch('/api/auth/csrf');
	const data = await response.json();



	@cursor-tasks.md Go through each story and task in the cursor-tasks.md file. Find the next story to work on. Review each unfinished task, correct any issues or ask for clarifications (only if absolutely needed!). Then proceed to create or edit files to complete each task. After you complete all the tasks in the story, update the file to check off any completed tasks. Run builds and commits after each story. Run all safe commands without asking for approval. Continue with each task until you have finished the story, then stop and wait for me to review.