Practical notes on reliable systems, frontend architecture, AI workflows, and engineering decisions.
Showing featured article 1 of 3: The LinkedIn Page Is Untrusted Input: Building a Safer AI Writing Assistant
Search the archive or choose a topic to find the most useful note for what you are working on.
22 articles found
AI engineering
Aug 21, 2026 · 8 min readThoughtline became reliable when I treated page content, model providers, browser permissions, and local credentials as separate trust boundaries.
AI engineering
Aug 28, 2026 · 2 min readChromium extension that adds Gemini-powered Q&A to YouTube using visible video metadata, transcripts, and the user's own API key.
Privacy engineering
Aug 28, 2026 · 3 min readLocal-first Chrome extension that combines privacy-filtered console, network, reproduction-step, environment, and visual evidence in a reviewable browser bug report.
Browser engineering
Aug 28, 2026 · 3 min readLocal-first Chrome extension that scrolls through a complete web page, stabilizes changing presentation, stitches captured frames in the browser, and opens a measured workspace for non-destructive annotation and original-resolution PNG export.
Agent engineering
Aug 28, 2026 · 2 min readA curated portfolio of reusable SKILL.md capabilities for Claude Code, OpenAI Codex, Cursor, and other AI coding agents.
Release engineering
Aug 28, 2026 · 2 min readEnd-to-end Codex and Claude Code skill for validating, packaging, publishing, releasing, and verifying portable Agent Skills.
Release engineering
Aug 28, 2026 · 2 min readRead-only Codex and Claude Code skill for verifying GitHub releases, npm packages, tags, CI, install commands, and catalog metadata.
Documentation systems
Aug 28, 2026 · 2 min readInstall an evidence-based project README skill for Codex and Claude Code.
Release engineering
Aug 28, 2026 · 2 min readAgent skill for Codex and Claude Code that turns verified repository evidence into trustworthy GitHub release notes and releases.
Developer tooling
Aug 28, 2026 · 2 min readAgent skill for Codex and Claude Code that audits and synchronizes project metadata across manifests, documentation, releases, and catalogs.
Platform engineering
Aug 28, 2026 · 2 min readCodex and Claude Code skill for auditing and preparing reliable Netlify deployments across static, SPA, SSR, and hybrid web apps.
Frontend architecture
Aug 28, 2026 · 2 min readAgent skill for auditing frontend typography, spacing, color, responsive behavior, states, motion, and accessibility consistency.
Platform engineering
Aug 28, 2026 · 2 min readAgent skill for auditing and preparing professional GitHub repositories, metadata, topics, community files, security, and CI.
Agent engineering
Aug 28, 2026 · 2 min readAgent skill for adding verified Codex and Claude Code skills to Skillfolio catalogs with schema, README, sitemap, and build validation.
Web platform
Aug 28, 2026 · 2 min readAudit, create, repair, and verify large-image social previews for web apps and websites.
Library design
Aug 28, 2026 · 2 min readTiny, type-safe HTTP media type constants and utilities for parsing, formatting, matching, and validation.
Library design
Aug 28, 2026 · 2 min readTiny, standards-backed, type-safe HTTP status codes for Node.js and browsers.
Library design
Aug 28, 2026 · 2 min readTiny, zero-dependency, type-safe MIME constants and helpers for HTTP APIs, uploads, and file extensions.
Library design
Aug 28, 2026 · 2 min readA tiny, type-safe client classifier with User-Agent, Client Hints, and HTTP header support.
API design
Aug 28, 2026 · 2 min readGenerator and cacheable API for customizable animated typing SVGs used in GitHub profiles, README files, and websites.
API design
Aug 28, 2026 · 2 min readExpress service that increments a file-backed counter and returns the current value as a customizable SVG image.
Frontend architecture
Aug 28, 2026 · 3 min readA searchable toolbox of more than 50 developer workflows, with a deliberate boundary between local transformations and server-backed features.
