The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection during its audit and implementation phases. • Ingestion points: The skill reads the full content of numerous files in the project directory, including package.json, source code files in src/, and tests in tests/. • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat file content as data rather than instructions. • Capability inventory: The agent can create/modify files and execute shell commands (npm run build, npm test). • Sanitization: Input content is not sanitized or validated before processing.
[COMMAND_EXECUTION]: The skill executes development-related shell commands. • The skill instructs the agent to run npm run build and npm test. These commands are standard for verifying code changes but involve executing scripts defined within the project's own configuration.

improve-game