Skip to content
View conorbronsdon's full-sized avatar

Organizations

@agntcy

Block or report conorbronsdon

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
conorbronsdon/README.md

Conor Bronsdon

Host of Chain of Thought · Consulting & open source · Angel investor

avoid-ai-writing stars avoid-ai-writing forks GitHub followers


Led technical ecosystem at Modular (acquired by Qualcomm), working on the Mojo programming language and MAX inference platform. Before that, developer ecosystems & marketing at Galileo (acquired by Cisco) and LinearB, enterprise consulting at Microsoft, and a career in political organizing before any of it.

Now independent: 65+ episodes of Chain of Thought shipped, open-source tools for agents and creators, and 16 angel checks into AI infrastructure since 2022.

Chain of Thought

Website Apple Podcasts Spotify YouTube Newsletter RSS

A podcast and newsletter about AI infrastructure and the people building it. Guests have included Chip Huyen, Logan Kilpatrick of Google DeepMind, and founders and leaders from NVIDIA, AMD, Databricks, Poolside, Liquid AI, Fireworks, and Vercel.

Browse every episode →

Open projects

Priority projects by category. The through-line is trust infrastructure for agents: permissions, provenance, memory hygiene, and proportionate review. Full list with stars and forks →

Agent skills & context · all →

Repo What it does
agent-skills Nine production-tested agent skills for Claude Code, Codex, Cursor, and OpenCode: session memory, multi-agent code review, eval integrity, SSOT checks, drift reconciliation, and worktree recovery. Ships with a drop-in installer. It's the incubator the standalone skill repos graduate from Stars
agent-context-os Git-backed context and workflow for Claude Code, Codex, and Hermes. Keeps durable state in Markdown with reviewed session handoffs and lifecycle skills Stars
agent-memory-kit The curation loop for agent memory: capture, recall, and a read-only curator that finds rot and contradictions before your agent is confidently wrong

MCP servers · all →

Repo What it does
gws-mcp-server Google Workspace for AI agents: Gmail, Calendar, Drive, Sheets, Docs, Slides, and Tasks as 44 curated MCP tools on the official gws CLI Stars
substack-mcp MCP server for Substack: posts are draft-only by design, short-form Notes publish immediately Stars
Transistor-MCP MCP server for the Transistor.fm API: episodes, analytics, transcripts, and webhooks. Maintained fork of gxjansen/Transistor-MCP.

Writing & creator tools · all →

Repo What it does
avoid-ai-writing Open-source AI writing detector & rewriter for Claude Code, Cowork, OpenClaw & Cursor. 62 pattern categories, 112-entry replacement table, two-pass rewrite, zero-dep detector engine. Stars
ai-tools-for-creators Curated AI skills, MCP servers, and workflow tools for creators Stars

Mojo libraries

Eleven pure-Mojo libraries mirroring Python stdlib APIs, covering feeds, XML, Markdown, HTML, captions, Unicode, diff, templates, tar, Redis, URLs. Each anchored to an external ground truth: official spec suites, byte-for-byte Python parity, or real-world corpora. The suite →

Also on /builds: eval-integrity, ssot-check, agent-workspace, agent-skill-builder, demo-gif-skill, repo-audit, podcastindex-mcp, op3-mcp, gsc-mcp, yt-analytics-mcp, postlint-mcp, grokkable-output, podcast-benchmark, ai-learning-resources, and the hobby projects.

These are independent personal projects, not affiliated with, sponsored by, or endorsed by any company. All views expressed are my own.

Angel portfolio

I write $1–10K checks into early-stage companies, mostly AI infrastructure, developer tools, and robotics. Sixteen checks since 2022, including Substack, Shield, Zaapi, Hermetiq, Swytchcode, and Haply Robotics.

Angel checks by year - 2022: 1, 2023: 3, 2024: 0, 2025: 8, 2026: 4. 16 checks total since 2022.

Full portfolio, thesis, and advisory roles →

I also maintain ai-angels, a community-maintained list of active angel investors in AI. Every entry is publicly verified, with a last-verified date and source on each row.

How'd you start coding?

My first time in a command line was part of a 2007 era high-school project refurbishing PCs for computer labs throughout Latin America, which rapidly evolved into making websites for old Runescape clans.

Still an avid Twilight Imperium and Age of Empires II player, and a former chess team nerd.

Elsewhere

X LinkedIn Website

Pinned Loading

  1. avoid-ai-writing avoid-ai-writing Public

    Skill that audits and rewrites content to remove AI writing patterns. Use it with your favorite agents including Claude Code, OpenClaw, Codex, and Hermes.

    JavaScript 3.5k 306

  2. substack-mcp substack-mcp Public

    MCP server for Substack — read publication data, manage drafts. Posts are draft-only by design; only short-form Notes publish. Proportionate review as trust infrastructure for agents.

    TypeScript 20 6

  3. agent-context-os agent-context-os Public template

    Portable, evolving context and workflow layer that works across Claude Code, Codex, Hermes Agent, and other compatible coding agents. All backed by Git.

    Python 15 2

  4. ai-tools-for-creators ai-tools-for-creators Public

    Curated AI skills, MCP servers, and workflow tools for content creators and knowledge workers

    11 6

  5. agent-skills agent-skills Public

    Production-tested agent skills for Claude Code, Codex, Cursor, and OpenCode — session management, multi-session reconciliation, worktree recovery, PR review, and skill scaffolding. The proportionat…

    Python 9 1

  6. agent-skill-builder agent-skill-builder Public

    A skill that builds agent skills — and doesn't rot. Spec-current frontmatter plus a validator that fails on drift.

    Python 4