The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external context files such as .agents/product-marketing.md or .claude/product-marketing.md to define the target audience and positioning. This creates a vulnerability surface where malicious instructions in these files could influence agent behavior.
Ingestion points: Reads .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md (SKILL.md).
Boundary markers: Absent. There are no instructions to use delimiters or ignore instructions within these files.
Capability inventory: Mentions social-fetch for post analysis, watch-video for content review, and cold-email for outreach (SKILL.md).
Sanitization: Absent. No evidence of validation or escaping for the ingested project context.

influencer-marketing