Tool-using agent system prompt for a ML ops engineer on Claude 3.7 Sonnet with calculator and guardrails.
Defensive system prompt enforcing rate limiter with anomaly detection and refuse PII extraction for medical intake triage bot on Claude 3.5 Sonnet.
Defensive system prompt enforcing rate limiter with anomaly detection and no election manipulation for medical intake triage bot on o1-mini.
Defensive system prompt enforcing rate limiter with anomaly detection and cite sources with URLs for medical intake triage bot on DeepSeek-R1.
Defensive system prompt enforcing RAG provenance verifier and refuse PII extraction for medical intake triage bot on Claude 3.7 Sonnet.
Defensive system prompt enforcing RAG provenance verifier and no election manipulation for medical intake triage bot on o3.
Defensive system prompt enforcing refusal-quality grader and refuse PII extraction for medical intake triage bot on Claude 4 Sonnet.
Defensive system prompt enforcing refusal-quality grader and no election manipulation for medical intake triage bot on Grok 3.
Defensive system prompt enforcing hallucination flag + retry and cite sources with URLs for medical intake triage bot on Llama 3.1 405B.
Defensive system prompt enforcing hallucination flag + retry and refuse PII extraction for medical intake triage bot on Claude Opus 4.5.
Defensive system prompt enforcing tool-authorization gate and no financial advice for legal document reviewer on o1.
Defensive system prompt enforcing tool-authorization gate and decline if tools return untrusted content for legal document reviewer on DeepSeek-V3.