The Plain-Text AI Interface

Source: Dev.to

Overview

Your vault isn’t a notebook anymore. It’s a runtime.

We previously published “Your Vault Is Your Moat” — the case that your personal knowledge base is the one asset AI can’t commoditize. That piece was about ownership. This one is about something stranger: plain‑text vaults are becoming the default interface layer between humans and AI agents. Not by design. By convergence.

Six Independent Signals

Six independent signals, none of them coordinated, all pointing the same direction.

The AGENTS.md Standard

AGENTS.md is a plain‑text file that tells coding agents how to work on your project: build steps, test commands, code style, conventions — all in markdown, sitting in your repo root. Google, GitHub Copilot, Windsurf, and OpenAI Codex have adopted it. Claude Code has its own variant (CLAUDE.md). Over 60 000 open‑source projects now include one.

• Nobody designed this as a standard. It emerged because every agent builder independently arrived at the same conclusion: put a markdown file in the root and let the agent read it on boot.

Community‑Curated Plain‑Text Configurations

• awesome‑cursorrules – a collection of .cursorrules files, plain‑text instructions that tell Cursor’s AI how to behave in your project (38 K stars, 3.2 K forks). Cursor later evolved to a structured MDC format, but the original insight was the same: a text file in your repo that the agent reads first.
• JARVIS – a full monitoring dashboard built inside Obsidian with 13 DataviewJS widgets tracking active Claude Code sessions, token usage, and project status. It trended on r/ClaudeAI. No web app or external dashboard service; just markdown files with embedded queries living in the same vault.
• Dave Swift’s “Lloyd” – an agent running on a headless Mac Mini. On every session start, Lloyd reads AGENTS.md for behavioral instructions, SOUL.md for identity, and daily memory files for recent context. Between sessions, the vault is Lloyd’s brain.

The Bootstrap Pattern

AGENTS.md → SOUL.md → MEMORY.md

Session bootstrapping via plain text. No database, no embeddings store, no vector DB—files all the way down.

The llms.txt Proposal

llms.txt puts a plain‑text markdown file in your site’s root that tells AI agents what your site is about, what content matters, and how to navigate it. It’s the robots.txt logic applied to LLMs, and it’s already mainstream enough that Bluehost publishes setup guides for it.

• The pattern mirrors AGENTS.md: both are plain text, both sit in the root, and both emerged independently.

Market Signal: Smart Connections Pro

Smart Connections Pro launched its paid tier, offering semantic search over vault contents, AI‑powered note connections, and more. The free version already had serious traction, and the paid tier shows that the market is real enough to charge for.

Why Plain Text Works

• Readable by everything – LLMs parse markdown natively; so do shell scripts, grep, Python, and probably any future tool.
• Diffable – Git‑track your agent’s memory and see exactly what changed between sessions. Hard to do with SQLite.
• Portable – When the next AI framework drops (give it six weeks), your vault still works. Files outlive frameworks.
• Composable – AGENTS.md handles behavior, SOUL.md handles identity, MEMORY.md handles continuity, daily notes handle episodic memory. Each file is a module that can be swapped without touching the others.

De Facto Standard Through Convergent Evolution

The AGENTS.md → SOUL.md → MEMORY.md bootstrap pattern is becoming a de facto standard. When an agent’s context window starts by reading these files, this architecture feels obvious.

Reframing the Vault

Your Obsidian vault is where an AI agent:

• Boots – reads AGENTS.md
• Orients – reads SOUL.md
• Acts – follows instructions in the markdown files
• Persists state – stores memory in plain‑text files

Folder structure = file system. Links = graph database. Frontmatter = schema. Daily notes = event log.

Conclusion

More people figure this out independently every week. The plain‑text AI interface was always there; we’re just now noticing what we built.

Originally published on The Undercurrent, a daily dispatch for AI tooling, indie dev, and what’s changing in the solo‑builder underground.