agent-protocol

原文涵盖的主题

• Inter-Agent Protocol
• Keywords
• Invocation Syntax
• Response Format
• Loop Prevention (Hard Rules)
• Rule 1: No Self-Invocation
• Rule 2: Maximum Depth = 2
• Rule 3: No Circular Calls

How C-suite agents talk to each other. Rules that prevent chaos, loops, and circular reasoning. agent protocol, inter-agent communication, agent invocation, agent orchestration, multi-agent, c-suite coordination, agent chain, loop prevention, agent isolation, board meeting protocol Any agent can query another using: Examples: Valid roles: ceo, cfo, cro, cmo, cpo, cto, chro, coo, ciso, gc, cdo, caio, cco, vpe | Role token | Advisor skill | |---|---| | gc | general-counsel-advisor (legal, contracts, term sheets) | | cdo | chief-data-officer-advisor (data strategy, training-data rights) | | caio | chief-ai-officer-advisor (AI strategy, evals, AI risk) | | cco | chief-customer-officer-advisor (retention, customer success) | | vpe | vpe-advisor (engineering delivery, DORA, eng hiring) | Invoked agents respond using this structure: Example: These rules are enforced unconditionally. No exceptions. An agent cannot invoke itself. Chains can go A→B→C. The third hop is blocked. If agent A called agent B, agent B cannot call agent A in the same chain. Each invocation carries its call chain. Format: Agents check this chain before responding with another invocation. When blocked: Return this ins

这里只展示节选，请以完整原文和当前要求为准。