The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external source material to extract stylistic features, which presents a surface for indirect prompt injection if the source data contains malicious instructions.\n
Ingestion points: The skill gathers samples from X (via x-api), articles, newsletters, emails, and website copy as defined in SKILL.md.\n
Boundary markers: The skill utilizes a structured VOICE PROFILE block and a specific schema in references/voice-profile-schema.md to delimit the extracted data, though it lacks specific 'ignore instructions' markers for the input phase.\n
Capability inventory: The agent uses x-api for data collection and the resulting profile influences downstream content generation and outreach tools like content-engine and lead-intelligence.\n
Sanitization: No explicit sanitization or filtering of the input source material is performed before analysis.

brand-voice