The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted TypeScript source files.\n
Ingestion points: The agent uses Grep and Read tools to scan *.test.ts, *.spec.ts, and *.test.tsx files.\n
Boundary markers: There are no explicit instructions to the agent to isolate or ignore instructions that might be embedded within the processed code files.\n
Capability inventory: The skill utilizes Write and Edit for file modification and Bash for command execution, creating a path for automated exploitation if malicious code is processed.\n
Sanitization: No sanitization or verification of the file content is performed before the agent acts upon the data.\n- [EXTERNAL_DOWNLOADS]: The skill initiates the installation of the @total-typescript/shoehorn package via npm. This is an established library from a well-known source intended for the specific purpose of the skill.\n- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute npx tsc --noEmit to verify type-checking results, which is a standard procedure in TypeScript development environments.

migrate-to-shoehorn