The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted content from the repository and conversation to generate a domain model. * Ingestion points: The skill reads repository files and thread discussion as specified in the 'Process' section. * Boundary markers: Absent; there are no instructions to use delimiters or to disregard commands embedded in the processed terminology. * Capability inventory: The agent has permissions to read files across the repository and write the 'UBIQUITOUS_LANGUAGE.md' file. * Sanitization: Absent; the skill does not specify any validation or filtering of content extracted from the codebase before writing it to a file.

ubiquitous-language