The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external sources and local configuration files that could be manipulated by an attacker to include malicious instructions.
Ingestion points: The instructions in SKILL.md direct the agent to read local context files like .agents/product-marketing.md and suggest gathering data from third-party review platforms such as G2, Capterra, and TrustRadius.
Boundary markers: There are no instructions for the agent to use specific delimiters or protective framing when processing this external content to distinguish it from its primary task instructions.
Capability inventory: The skill is primarily focused on text generation and architectural planning; however, it references other skills like programmatic-seo which may have broader capabilities that could be influenced by injected data.
Sanitization: The research process does not include steps to validate or sanitize external review data or local marketing context for adversarial patterns.

competitors