multi-agent-debate vs a2a_call_agent
| multi-agent-debateio.github.moralito311-andr/andreax | a2a_call_agentio.github.Tim-Projekt/agent-router | |
|---|---|---|
| Behaviour score | 100.0 | 100.0 |
| Measurement tier | shallow | shallow |
| Trials | 1 | 1 |
| Schema integrity | 100 | 100 |
| Transport | remote | remote |
| Auth | self-provisionable | self-provisionable |
| Side effects | unclassified | unclassified |
| Input schema | not declared | not declared |
| Output schema | not declared | not declared |
| Measured | 2026-08-31 | 2026-08-25 |
Every row is a measured fact or a declared property — the full per-probe evidence behind each score is on that tool’s own report card. Deep-tier and shallow-tier results are never averaged into one number; see the methodology.