The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions to use shell-based tools like rg (ripgrep) for project discovery and npm test for verifying implementation logic. These commands are localized to the development workflow and serve the primary purpose of the skill.
[EXTERNAL_DOWNLOADS]: References are made to standard Node.js package management workflows (e.g., npm test, npm run build) and dependencies on well-known libraries like @nestjs/common, typeorm, and rxjs.
[DATA_EXFILTRATION]: The skill focuses on database transaction logic and does not contain patterns for accessing sensitive files (e.g., SSH keys, cloud credentials) or transmitting data to external domains.
[PROMPT_INJECTION]: The instructions do not contain markers for overriding agent behavior or bypassing safety filters. All instructions are within the scope of the stated technical purpose.
[DATA_EXFILTRATION]: Regarding Indirect Prompt Injection (Category 8):
Ingestion points: Reads target project source files, package.json, and tsconfig.json during the discovery phase.
Boundary markers: Not explicitly defined in the provided file reading instructions.
Capability inventory: Includes the ability to execute discovery commands (rg) and build/test scripts.
Sanitization: Not explicitly implemented within the skill logic. This represents a standard surface for developer-centric skills processing codebase content.

nestjs-typeorm-transactional