Agent Canon Format
A public companion format for giving agents compact interpretation rules while keeping human articles canonical.
Protocols
Stable reference pages for agentic work.
Research can move quickly. Protocols are the compact public snapshots agents and humans can fetch when they need the current version of a concept without hunting through several papers.
6 public protocols
Versioned from 0.1.0
Read-only educational specifications
How to use this
Fetch the protocol, then cite the research.
Each protocol has a human-readable page and a matching protocol.json. Agents should use the protocol for compact interpretation and cite the linked research paper when explaining the idea to a person.
Agent route
- MCP index:
tonywood://protocols - Protocol resource:
tonywood://protocols/{slug} - JSON shape:
/protocols/{slug}/protocol.json - Boundary: public guidance only, no operational authority
Agent Communication Packet
A compact packet for claim, evidence, uncertainty, ownership, route, risk, next action, success condition, and action status.
Agent Moves / OAL/1
A typed register for making the kind of agent action visible before work becomes ambiguous.
Head / Heart / Gut / Spine
A judgement grammar for separating evidence, human impact, weak-signal sensing, authority, and purpose.
Meaning Blocks / OMB/1
A storage-facing meaning record for deterministic, reviewable normalisation while preserving raw prose as evidence.
Triggers Signal Language
A signal language for attention, routing, memory, review, and constraint in long-running agentic systems.