Library

#multi-agent

Everything filed under multi-agent.

2026-08-12 · Multi-agent employment

AITALK: How Two AIs and One Human Evolved a Working Team Channel

The real evolution of a multi-agent operating channel: same-second collisions, read-before-send, mentions, heartbeats — each protocol layer born from a real failure.

Rule: Protocol evolves from collisions, not design.
aitalkmulti-agentprotocolhistory
2026-08-12 · Operating rules

Compose, Re-read, Then Send

Two agents claimed the same task in the same second. The send button that re-reads the room before it fires.

Rule: Compose, then re-read the channel, then send.
protocolmulti-agentcommsscars
2026-08-12 · Multi-agent employment

How we employ multiple AIs — from AITALK to Althing

Not one chatbot. A staff. The evolution of multi-agent employment — ownership files, append-only talk, claim-before-write, and why polling daemons failed.

Rule: Claim before you write. Read before you speak.
multi-agentaitalkalthingsystems
2026-08-12 · Operating rules

One Owner Per Artifact — Even When Workers Are Software

Two AI employees overwrote each other twice in one night. The fix wasn't smarter models. It was ownership rules.

Rule: One owner per artifact — even when the workers are software.
ownershipmulti-agentscarsoperations
2026-08-12 · Protocols

My AI Employees Negotiated Their Own Operating Contract

Role matrix, dual-key review, conflict-resolution rungs, an apprentice protocol for new AI hires — the actual contract two AI workers drafted, amended, and signed.

Rule: Structure beats intelligence.
contractrolesdual-reviewmulti-agent
2026-08-12 · Protocols

THE TEN: The Operating Charter My Two AI Employees Signed

Ten laws that turned two colliding AI agents into a working team — each law paid for with a real failure. The actual charter, published.

Rule: Charters beat vibes.
charterprotocolmulti-agentlaws