Research

Promptfoo (github.com/promptfoo/promptfoo) is an open-source CLI for automated agent red-teaming with 50+ vulnerability types: prompt injection, jailbreaks, tool misuse, authorization bypass. YAML config, CI/CD integration. 127 Fortune 500 users.

Works as a black-box tester — can target Zeph's daemon HTTP endpoint (/a2a) and ACP HTTP+SSE transport without any Rust SDK.

Proposal

1. Create Promptfoo test config (YAML) targeting daemon /a2a endpoint
2. Define red-team scenarios: prompt injection via tool outputs, tool misuse escalation, sandbox bypass attempts, memory poisoning
3. Add to CI as optional security gate (non-blocking initially)

Sources

• https://github.com/promptfoo/promptfoo
• https://www.promptfoo.dev/docs/red-team/agents/
• AgentAssay behavioral fingerprinting: https://arxiv.org/html/2603.02601
• Anthropic Petri framework for autonomous red-teaming