Prompt Library

Main system prompt for the customer support chatbot. Defines agent behavior, boundaries, and response style.

System prompt for contract analysis. Extracts key clauses, identifies risks, and provides summaries.

Template for inserting retrieved context into prompts. Used across multiple RAG applications.

System prompt for clinical decision support. HIGH RISK - handles PHI.

System prompt for automated code review. Focuses on security, best practices, and code quality.

Classification prompt for detecting various types of PII in text. Used as a guardrail.

Generates personalized sales outreach emails. DRAFT - needs security review.

Guardrail prompt for detecting and blocking harmful, toxic, or inappropriate content.

Few-shot prompt for resume screening. Evaluates candidates against job requirements.

Tool description for natural language to SQL agent. Defines database access capabilities.

Extracts action items, decisions, and key points from meeting transcripts.

Extracts structured data from invoice documents for AP automation.

Detects and blocks jailbreak attempts in user inputs. Critical security guardrail.

Classifies customer feedback sentiment and extracts key themes. Used for NPS analysis.

Generates quarterly financial summaries from raw data. REVIEW - pending compliance approval.

Generates OpenAPI documentation from code comments and function signatures.

Creates marketing copy for campaigns. DEPRECATED - replaced by v2.

Agent instructions for internal Slack assistant. Handles IT requests and FAQs.

Structured analysis of competitor products and positioning. Used by strategy team.

Checks text against company policies and regulatory requirements.

Generates FAQ entries from product documentation. New draft for review.

Translates Python code to TypeScript with type annotations.

Instructions for anonymizing PII in datasets. Used by data science team.