Operating manual
Tag: Multi-Agent Systems
A focused reading path for Multi-Agent Systems: related field notes, evidence trails, and operating questions from the archive.
- scoring-definition-is-the-lever-in-agent-evals.md
Scoring Definition Is the Lever in Agent Evals
A 24.0%-57.0% consensus swing on identical held-out data is a property of the scoring definition, not the agents. Lock the definition before reading the number.
open artifact → - agent-depth-trades-yield-for-cheaper-context.md
Agent Depth Trades Yield for Cheaper Context
Depth is not automatically cheaper context. Treating finding yield, root-context exposure, and dollars per finding as three independent axes exposes the real trade, and a per-tier alignment check prevents promoting another tier on paper numbers alone.
open artifact → - the-missing-check-in-agent-handoffs.md
The Missing Check in Agent Handoffs
A practical two-sided contract for checking whether one agent's completed work leaves state the next agent can actually use.
open artifact → - the-authority-envelope-every-agent-handoff-should-carry.md
The Authority Envelope Every Agent Handoff Should Carry
Agent handoffs do not become trustworthy because two systems can talk. They become trustworthy when every handoff carries authority, state, evidence, risk, and the next allowed action.
open artifact → - agent-teams-need-governance-before-they-need-another-protocol.md
Agent teams need governance before they need another protocol
Protocols make agent collaboration possible. Governance makes it safe enough to delegate: authority, state boundaries, evidence, review, and rollback.
open artifact → - the-agent-stack-is-getting-protocols-before-it-gets-governance.md
The Agent Stack Is Getting Protocols Before It Gets Governance
MCP, A2A, and agent SDKs make agents easier to connect. Production trust depends on the control layer above them: authority, review, evidence, recovery.
open artifact →