The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external information from local context files and web search results to generate product listings.
Ingestion points: Reads contextus.md from the root directory and instructs the agent to search the web for missing product details (pricing, features, reviews).
Boundary markers: None explicitly defined in the skill instructions to delimit untrusted web content.
Capability inventory: The skill uses text generation and web search capabilities. It does not invoke dangerous tools such as shell execution, filesystem writes, or network exfiltration.
Sanitization: No explicit sanitization or filtering of the ingested data is mentioned.
Conclusion: This is a standard functional requirement for the skill's purpose (marketing copy generation) and does not present an exploitable risk in the absence of high-privilege tools.
[SAFE]: The skill references an external documentation link for additional guidance (alignify.co). This is documented neutrally as a resource for directory submission strategies and does not involve code execution or suspicious redirects.

directory-submission