The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, such as web search results from community forums (Reddit, Quora) and user-provided sales transcripts or SEO exports. While this represents a vulnerability surface for indirect prompt injection, it is a necessary component of the skill's research function and does not lead to unsafe actions.\n
Ingestion points: SKILL.md identifies Reddit/Quora search results, competitor analysis, and customer call transcripts as primary data sources.\n
Boundary markers: The instructions do not explicitly define delimiters for external data but use specific analysis templates to scope the agent's interaction with the inputs.\n
Capability inventory: The skill utilizes standard informational tools (file reading and web search) without granting access to sensitive system resources or arbitrary code execution.\n
Sanitization: No specific content sanitization or instruction-filtering is mentioned.\n- [SAFE]: The skill instructions and referenced guides for distribution and CMS usage focus entirely on legitimate marketing and SEO strategy. The external references and relative links within the repository follow standard documentation practices and do not exhibit signs of obfuscation, exfiltration, or persistence mechanisms.

content-strategy