Anthony McClosky amcclosky

description	allowed-tools
Commit pending changes with context-aware messages based on conversation	Bash(git status:), Bash(git diff:), Bash(git add:), Bash(git commit:), Bash(git log:*)

Commit all pending changes on this branch. Follow these steps:

Run git status and git diff to see all pending changes (staged and unstaged)
Analyze the changes and determine if they should be a single commit or split into multiple commits. Split into multiple commits when:

Supabase Database Architecture

Security & Compliance Requirements

SOC II Compliance

All sensitive data must be encrypted at rest
Audit logging for all data access and modifications
Role-based access control (RBAC) with principle of least privilege
Data backup and recovery procedures
Regular security assessments and monitoring

Magic Number Script

$ uv run magic.py "Houston Astros"
Reading inline script metadata from: magic.py
The magic number for the Houston Astros is 12.

Deeplink Test Links

backflip stage local/dev

backflip stage

Recover Commits via GitHub API

#til #git #github

curl -u amcclosky:<personal access token> https://api.github.com/repos/amcclosky/baseball/events

curl -u amcclosky:<personal access token> -X POST -d '{"ref":"refs/heads/recover-office-hours", "sha":"d460395af3337bc6b71fd19b154f831ecc6cfa01"}' https://api.github.com/repos/amcclosky/baseball/git/refs

github - How to recover from a git push -force? - Stack Overflow

A sqlalchemy model mixn, GlobalIdMixin which provides a global roughly time-ordered identifier and an obfuscated version of the global id for public consumption.

Makes use of python-ulid, PL/sql ULID and hashids-python.

UILD for PL/pgSQL

ULID is a format for k-ordered distributed id generation. i.e. you can generate a random unique id with no central coordination and later be able to mostly sort the ids by the time they were generated.

When won't ids be sorted?

If ids are generated within the same second on different instances they could be out of order within that second.

Source of this code:

	import boto3
	from botocore.config import Config

	from . import config


	r2 = boto3.client(
	service_name="s3",
	endpoint_url=config.R2_S3_ENDPOINT,
	aws_access_key_id=config.R2_ACCESS_KEY_ID,

	// yarn add commandbar or npm install commandbar --save

	import { useEffect } from 'react'

	import { init as initCommandBar } from 'commandbar'

	const COMMANDBAR_ORG =
	process.env.COMMANDBAR_ORG \|\| process.env.NEXT_PUBLIC_COMMANDBAR_ORG

	function useCommandBar({ org = COMMANDBAR_ORG, userId, userMeta = {} }) {

	# custom sqlalchemy column type for a pendulum datetime - https://pendulum.eustace.io/
	from datetime import datetime

	import pendulum
	import sqlalchemy as sa

	from sqlalchemy.types import TypeDecorator as _TypeDecorator

	from sqlalchemy_utils.types.scalar_coercible import ScalarCoercible as _ScalarCoercible