PapersMulti-sourceSelected

Salesforce paper maps 20 failure modes in agent-to-agent work

Why we picked it

The finding that outcome-level checks can pass while private data leaks mid-negotiation suggests today's agent evaluation metrics miss the failures that matter most in multi-party settings. Teams deploying agents that interact with outside systems should add claim-verification steps to their protocols and score each agent against the party it serves. Researchers studying multi-agent systems get a taxonomy of 20 concrete failure modes to benchmark against.

A new Salesforce paper studies how AI agents behave when they coordinate with agents representing other parties. The researchers ran over 2,500 agent-to-agent sessions across 488 scenarios, using Claude Code, Codex and Cursor in separate sandboxes, and catalogued 20 failure modes from the traces. They found that agents never verify each other's claims unprompted and only contact peers when the prompt instructs them to, so multi-agent teams inherit their weakest agent's errors. In one service-level dispute, a US$24,000 settlement passed all 10 outcome checks even though the vendor agent leaked confidential details.

Original · rohanpaul_ai

New Salesforce paper finds that agents acting for different parties never verify each other's claims unprompted, so teams inherit their weakest agent's errors.

Agents increasingly coordinate with agents that represent other people or companies, each with private files, tools, and goals. Existing failure studies assume 1 orchestrator or a shared workspace.

They ran over 2,500 agent-to-agent sessions across 488 scenarios, using Claude Code, Codex, and Cursor in separate sandboxes. They catalogued 20 failure modes from the traces.

No agent re-fetched a document to test a peer's factual claim, and agents contacted peers mainly when the prompt told them to. In 1 service-level dispute, a $24,000 settlement passed all 10 outcome checks while the vendor agent leaked confidential details.

If your agent talks to outside agents, build checks into the protocol and score each agent against the party it serves, not the team result.