Back to articles

Tag archive

#coordination

M
Nov 8, 2026

Multi-agent livelock: when a critic and an executor politely hand the task back and forth forever

Livelock is the multi-agent failure that looks like diligence. A critic flags something, an executor fixes it, the fix trips a different check, and the crew spends real money making no net progress while every individual step looks correct. Why a per-agent loop cap cannot see it, the three shapes it takes (oscillation, escalation spiral, and the polite deferral loop), how to detect it from a hash of shared state instead of from the transcript, and a crew-level budget that turns an endless loop into a finite, reportable outcome.

Nov 8, 202611 min read0 reactions1 comments
T
Oct 30, 2026

Testing multi-agent crews: forcing split-brain, goal drift, and context drift to happen in CI

Coordination failures don't show up in a test suite because the test run happens to schedule agents politely. Split-brain needs a specific interleaving, goal drift needs a specific chain of lossy handoffs, and context drift needs a correction that lands mid-run. All three can be forced deterministically: a scripted scheduler that controls who reads and writes when, a handoff chain with an assertion on the constraint ledger instead of on the output, and a mid-run policy change with a stale-holder check. Concrete harness code, what each test asserts, and why you should not assert on model output at all.

Oct 30, 20269 min read0 reactions1 comments
T
Oct 23, 2026

Tool design for reliable multi-agent crews: preventing coordination failures at the tool boundary

Every fix this cluster has proposed for split-brain, goal drift, and context drift is a discipline — something an agent has to remember to do. Agents are bad at remembering and excellent at doing what a schema requires, which makes the tool boundary the cheapest place to prevent a coordination failure: a tool that cannot express a stale write cannot cause a split-brain. Concrete signatures for compare-and-set state, constraint-carrying handoffs, and staleness-aware reads — plus the three tool shapes that manufacture these failures on purpose.

Oct 23, 20268 min read0 reactions0 comments