The Agent Skills Directory

[PROMPT_INJECTION]: The skill is configured to ingest untrusted data from the local environment, creating a surface for indirect prompt injection.
Ingestion points: The skill instructions in SKILL.md direct the agent to read context from files such as .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md.
Boundary markers: There are no explicit boundary markers or instructions provided to the agent to disregard potentially malicious instructions contained within these context files.
Capability inventory: The skill generates advertising copy, strategy documents, and interfaces with advertising platform MCP tools (Google Ads, Meta, LinkedIn, TikTok).
Sanitization: The skill does not define any sanitization or validation procedures for the content retrieved from these context files.
[SAFE]: The skill includes technical implementation guides for advertising tracking pixels that reference well-known and trusted services.
The references/conversion-tracking.md file contains code snippets for Google Tag Manager, Meta Pixel, LinkedIn Insight Tag, and TikTok Pixel. All associated URLs point to official, well-known domains (e.g., googletagmanager.com, facebook.net, licdn.com, tiktok.com) and are intended for standard user implementation.

ads