The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external, public-facing sources that could contain malicious instructions designed to manipulate the agent's reporting or project state.
Ingestion points: The skill retrieves content from business reviews and profile questions using the get_business_reviews and get_google_business_questions tools as specified in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard or escape potentially malicious text embedded within reviews or questions.
Capability inventory: The agent has the capability to write durable project data using update_project_context and generate HTML-based reports via the seo-report skill, which could be influenced by injected instructions.
Sanitization: No sanitization or validation procedures are defined to filter or clean external content before it is processed or stored.

local-seo