The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection through its automated type-fixing command.
Ingestion points: The agent is instructed to read project metadata from package.json and tsconfig.json, as well as source code files using Read, Glob, and Grep (commands/typescript-fix.md).
Boundary markers: The instructions do not define delimiters or provide warnings to the agent to disregard natural language instructions that might be embedded within the project's code or configuration files.
Capability inventory: The skill utilizes powerful capabilities including Bash for command execution and Edit/Write for file modification across the project directory (commands/typescript-fix.md).
Sanitization: No explicit sanitization or validation of the ingested project data is performed before it is used to determine the execution of shell commands.
[COMMAND_EXECUTION]: The skill facilitates the execution of local development commands.
The typescript-fix command instructs the agent to execute project-defined scripts (e.g., npm run typecheck) and the TypeScript compiler (npx tsc) using the Bash tool. While these are standard developer workflows, they represent a path for executing code defined within the analyzed project.

typescript-expert