Diff two A2A AgentCard documents — classify changes (autonomy escalation, memory escalation, tool/model removal, incident-response URI removal, etc.), surface a breaking flag. The agent-cards counterpart to mcp-tools-diff. Library + CLI.
Diff two A2A Agent Card documents and classify the changes. The agent-cards counterpart to . Status: v0.1.0 — Node 20/22 supported, library + CLI. Agent cards declare an agent's autonomy level, memory persistence, models, tools, refusal taxonomy, evaluations, and incident-response posture. Many of those are load-bearing for procurement and security review. When the card changes between two…
Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts — it cannot prove the absence of malicious code.
Diff two A2A Agent Card documents and classify the changes. The agent-cards counterpart to . Status: v0.1.0 — Node 20/22 supported, library + CLI. Agent cards declare an agent's autonomy level, memory persistence, models, tools, refusal taxonomy, evaluations, and incident-response posture. Many of those are load-bearing for procurement and security review. When the card changes between two versions, which changes break the trust assumptions downstream operators already approved? This library…