The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted user-provided source code which could contain malicious instructions intended to influence the agent's behavior.
Ingestion points: The skill reads user code files, package.json, and directory structures using the Read, Glob, and Grep tools.
Boundary markers: There are no explicit instructions for the agent to treat code comments or strings as untrusted data or to ignore embedded instructions within the source code.
Capability inventory: The skill possesses the Edit capability to modify local files and a restricted Bash capability to execute the tsc compiler.
Sanitization: No explicit sanitization or validation is performed on the content of user-provided files before they are processed or used as input for the compiler.

typescript-engineer