The Agent Skills Directory

[PROMPT_INJECTION]: The skill relies on analyzing data from untrusted external sources, such as customer support tickets and user surveys, which introduces a surface for indirect prompt injection attacks where malicious instructions could be embedded in the data.\n
Ingestion points: External data enters the agent context through discovery interviews, surveys, sales calls, support tickets, and reviews as described in the prioritization section of SKILL.md.\n
Boundary markers: The instructions do not define boundary markers or clear delimiters to isolate the ingested data from the agent's core instructions.\n
Capability inventory: The skill has access to powerful tools including Write, Edit, and Bash as defined in the frontmatter.\n
Sanitization: No sanitization, validation, or filtering procedures for the external content are specified in the skill's instructions.

pm-requirements