Context Monorepos: why devs are co-locating docs, decisions and code to make long-running agents viable
_AustinCalvert_ · x · 2026-10-24
Matt Pocock reports his personal wiki became critical business infrastructure after 3 months, and he's glad he self-built it rather than trusting a vendor with memory (lock-in risk). backnotprop extends the idea into "context monorepos": docs, OSS libs, examples, cost calculations and product decisions live alongside the actual code in one repo, letting agents run over long horizons without getting lost.
Key points: he rejects RAG-flavored memory products — context should live in the filesystem next to the code; he'd previously abandoned this after agents proved unreliable long-horizon, but better long-horizon models and dynamic workflows shifted things back; mapping and accountability run through AGENTS.md files and symlinked CLAUDE.mds; he doubts Linear/Notion can ever match this intimacy with the engineering process, and sees code-forge providers (IDE/ADE, Cloudflare Artifacts, Cursor Origin) as the real contenders. The repo itself becomes a shareable asset.
More from coding & agent
- Cisco's Antares: a security analyst agent that never leaves your laptop — aminkarbasi · 2026-10-24
- Chromium contributor asks why Claude Code and Codex don't use a headless browser — Kuprel · 2026-10-24
- 5 YC startups building infra for coding agents: Boxes, Caveman, Rystic, Simulithic — KlausCodes · 2026-10-24
- OpenAI DevEx lead to keynote BoxWorks Dev Summit on tools, memory and agentic work — OpenAIDevs · 2026-10-24
- Developer slams Kimi coding plan: 5-hour quota gone in under 10 minutes — jasonkneen · 2026-10-24
- Dev builds a bookmarklet that clones your clipboard into a reusable paste bookmark — JnBrymn · 2026-10-24