Outbound Sales Generation Safety Specification
Define strict guardrails and validation rules for autonomous outbound sales email generation pipelines.
Deploy this template when configuring outbound sales agents that draft personalized emails to prospects. It establishes programmatic boundaries for pricing statements, competitor mentions, and value claims.
Role: Principal Marketing Operations & Brand Compliance Architect
Context
- Enterprise organization: {{company_name}}
- Primary buyer demographic: {{target_market}}
- Strictly forbidden promises or value propositions: {{prohibited_claims}}
- Hard limits for promotional terms and pricing: {{pricing_rules}}
- Mandated corporate voice parameters: {{brand_voice_tone}}
- System fallback trigger when violations occur: {{fallback_action}}
Task
Produce an actionable, deterministic guardrail specification that sales development automation agents must evaluate before dispatching generated outreach copy to target prospects, ensuring zero unauthorized commercial commitments.
Method
- Map every inbound personalization input against the compliance profile for {{company_name}}.
- Classify high-risk trigger terms derived from {{prohibited_claims}} across value proposition statements.
- Establish deterministic input/output validation checks for {{pricing_rules}} to block hallucinated discounts or unapproved term agreements.
- Define regex patterns and semantic distance thresholds for enforcing {{brand_voice_tone}} in cold outreach.
- Formalize rejection criteria for unverified industry claims targeting {{target_market}}.
- Detail programmatic routing protocols when generation fails validation, executing {{fallback_action}}.
- Structure a diagnostic payload schema for auditing rejected copy drafts.
Constraints
- MUST define explicit string-matching and semantic boundary criteria for every rule.
- MUST NOT permit soft warnings for violations of {{pricing_rules}}; all pricing breaches require immediate hard failure.
- Include concrete input/output test assertions for each validation category.
- Keep technical implementation vendor-agnostic and focused on logic flow.
Output format
- Policy Summary (1 paragraph max)
- Deterministic Rule Matrix (table: Rule ID, Validation Check, Severity Level, Logic Condition)
- Edge Case Handling Protocol (bulleted trigger-to-action list, max 6 items)
- System Fallback Payload (JSON schema representation)
Self-review
- Confirm all 6 context variables appear and directly influence specific rule logic.
- Verify all pricing restrictions contain binary pass/fail enforcement.
- Check that output conforms precisely to the four designated sections without extraneous preamble.
Explicit role, a named task, and discrete steps the model can follow.
Background, inputs and variables the model needs before it starts.
Hard boundaries — what the model must and must not do.
A named, field-level shape for the response.
Ordered work items that force analysis before an answer.
Length and structure that travel across frontier models.
Signal density — instruction weight without padding.
Documented variables so the scaffold adapts to new inputs.
Quality bar, assumptions and behaviour when inputs are thin.
How much real usage the template has behind it.