The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external 'available evidence' such as prospect data and market signals, creating a surface for potential indirect prompt injection attacks.
Ingestion points: The available evidence and account or segment inputs defined in SKILL.md and references/procedure.md.
Boundary markers: The skill employs a structured tagging system (observed, asserted, hypothesis, unknown) to categorize information and maintain context boundaries.
Capability inventory: The workflow utilizes the omh CLI for runtime recording and supports subagent delegation for specialized tasks.
Sanitization: It includes mandatory procedures like sales_account_evidence_check and sales_sequence_eligibility_check to validate that all claims are supported by approved source material.

omh-sales-development