The Agent Skills Directory

Indirect Prompt Injection Surface: The skill processes incoming cease-and-desist letters which are external, untrusted documents. This creates a potential vulnerability where malicious instructions embedded in the correspondence could influence the agent's behavior during the triage process.
Ingestion Points: The 'Receive' mode ingests external content from incoming letters, which may be provided as pasted text or via file paths (e.g., ~/Downloads/incoming-cd-acme.pdf).
Boundary Markers: While the skill is instructed to perform specific extraction and assessment tasks, there are no explicit instructions to use boundary markers or delimiters that would isolate the untrusted content from the agent's internal logic.
Capability Inventory: The skill has access to file-writing tools for generating .docx files, matter management functionality, and is directed to retrieve information from external research sources for counterparty diligence.
Sanitization: The instructions do not specify any sanitization, filtering, or validation steps for the content extracted from the incoming legal letters before it is analyzed.

cease-desist