#spec

Open, MIT context format for AI coding agents: centralized .aigx/ rules + a per-file boundary index that targets the right rules, forbidden imports, and gotchas to each file. The only context format validated in a controlled benchmark. Tool-agnostic; nothing injected into your source.

Protean-kit protocol ingredient for agent messaging. Dense, checkable messages for AI agents. AI agents talk to each other in English, usually wrapped in JSON. SYM-2P replaces the English with symbols: roles, actions, and states become numbers, fields become positions in a line, and payloads become s-expressions, a Lisp notation for nested structure. Every word an agent writes costs tokens, and tokens cost money and time. English is also ambiguous: no program can check a sentence the way it checks a number. Symbols are cheaper and checkable. One brief packet is a 101-byte line where the JSON took 481.