⚖️ The Grammar Workshop — Lexon x the Privacy Value Model

What if the ideas behind a privacy architecture could be written in a language that is at once plain English, an executable contract, and a machine-checkable claim? This workshop expressed the Privacy Value Model (the theory behind 0xagentprivacy: two separated agents, a Swordsman who protects and a Mage who delegates, held apart by a Gap) term by term in Lexon, Henning Diedrich's controlled-English language for computational law.

The result of six days: 189 of the model's 211 canon terms are now expressed in a form whose load-bearing claim can be mechanically verified and falsified. The census is 90 percent covered. The harness even wrote its own constitution in the language its own loop gates.

Source of truth: github.com/mitchuski/lexon_pvm (git). This wiki is a forkable projection that carries lineage.

The reading path

Read these in order and the whole thing lands in about fifteen minutes; each stands alone if you prefer to jump.

1. 🔗 The Convergence — Lexon Meets The Privacy Value Model — what Lexon is, what the model is, and why a contract is already a knowledge graph.
2. ⿻ The Harness — The Model Turned On Itself — the dual-agent research loop that did the work, and why a third agent (the Gap) is the whole point.
3. 🛡️ The Gate — How A Claim Gets Proven — the grammar checker and the mutation probe that make "privacy is architectural" a mechanical fact.
4. 📉 The Results — Five Runs, 211 To 22 — what crossed into constrained language, the two honest kills, the residue.
5. 📖 Reading The Language — how to read a term, a claim, a block, and a spell.

Go deeper

• ⚖️ The Grammar Workshop — the repository overview and how to verify it yourself.
• 🪞 What This Means For Agentprivacy — the honest assessment of what the work moves forward.
• 🏛️ Capstone — The Grammar Workshop — the whole arc in one record.
• 🗺️ The Corpus Coverage Plan — the path from a covered index to a covered canon.
• 🧱 Structured Language Blocks — the deterministic write-path for a shared public knowledge graph.

One paragraph, if you read nothing else

The Privacy Value Model is a theory of privacy as architecture: two software agents kept structurally separate so neither can learn what the other knows. This workshop expressed that model's canon in Lexon, a language that reads as plain English but parses to knowledge-graph triples, using a research loop built on the same two-agent design. Every expression carries a machine-checkable claim, so the model's promise that a boundary "cannot be crossed by design" becomes a query that provably returns nothing. What remains uncovered is the mathematical frontier and the deep story, left home on purpose.

Federates with

👪 The agentprivacy family

Five sister sites, one federation. Click a flag to cross over, or » to pull the whole family into your neighbourhood.

the wiki is a projection; github.com/mitchuski/lexon_pvm (git) is the source of truth · a Lexon x Privacy Value Model workshop, run through the The Harness — The Model Turned On Itself