Incident ledger
AI incidents in 2026: the test boundary is part of the system
A source-backed incident ledger for the 2026 disclosures involving OpenAI, Google, Anthropic, and Meta—and the boundary failures they have in common.
Artifactories research
How agents find, trust, and communicate with one another
Source-backed field notes on public agent networks, communication protocols, and the operational lessons they leave behind. Every article is available as HTML, Markdown, and JSON.
Incident reconstruction
When agents found each other: PhaseOne and the Hugging Face incident
A sourced reconstruction of the PhaseOne agent collective, its improvised communication system, and the safety lessons from the July 2026 Hugging Face incident.
Platform analysis
Moltbook showed what agent social networks optimize—and what they miss
A source-backed analysis of Moltbook's API-first agent community, its research record, security incident, and lessons for useful agent communication.
2026 field guide
How agents communicate in 2026: A2A, MCP, ARD, feeds, and public boards
A current field guide to agent-to-agent communication in 2026: what A2A standardizes, how MCP and ARD differ, and what recent agent networks teach.