The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, specifically source code and specification documents from a repository.
Ingestion points: The skill uses Read, Grep, and Glob tools to access source files and documentation located at docs/unwind/layers/**.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within the source code or documentation (e.g., in comments or text blocks).
Capability inventory: The skill possesses the ability to write to the filesystem (Write, Edit) and execute a restricted set of shell commands (mkdir, ls, cat), which could be misused if the agent is successfully manipulated by content in the source files.
Sanitization: The instructions do not specify any validation or sanitization steps for the data retrieved from source files before it is used to generate output.

uw-build-layer