The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that automatically ingests external configuration data, creating a surface for instruction override.\n
Ingestion points: SKILL.md directs the agent to read .agents/product-marketing-context.md or .claude/product-marketing-context.md to gather initial task context.\n
Boundary markers: Absent. The instructions do not specify any delimiters or safety prompts to isolate the ingested context from the agent's core instructions.\n
Capability inventory: The skill provides detailed playbooks for automating CRM operations in HubSpot and Salesforce and connecting workflows via Zapier. These capabilities could be misdirected by a poisoned context file.\n
Sanitization: Absent. The skill provides no guidance on validating or escaping the content of the markdown files before ingestion.

revops