The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Shell tool to execute various commands including pnpm run lint, pnpm exec tsc, and multiple grep patterns to analyze the codebase for quality issues.
[COMMAND_EXECUTION]: Employs node -e to execute inline JavaScript scripts for complex tasks such as file system traversal, line counting, and naming convention validation.
[COMMAND_EXECUTION]: Includes a PowerShell script block designed to detect consecutive lines of commented-out code, which is executed via the Shell tool when supported by the environment.
[DATA_EXPOSURE_AND_EXFILTRATION]: While the skill analyzes the codebase for the presence of the Cognite SDK client (new CogniteClient), this is performed locally via grep to ensure proper dependency injection and does not involve transmitting credentials externally.
[PROMPT_INJECTION]: The skill includes an instructional surface where it reads and processes source code containing comments and string literals. This represents an indirect prompt injection surface as an agent might encounter instructions embedded in code comments, though the skill's logic is primarily focused on pattern matching and reporting rather than executing instructions found within the data.
Ingestion points: Reads source code files (.ts, .tsx) from the src/ directory using Grep and Read tools.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded instructions within the source files.
Capability inventory: The skill has access to Shell, Write, Read, Glob, and Grep tools, allowing it to modify files and execute commands based on its analysis.
Sanitization: There is no evidence of sanitization or filtering of the source code content before it is processed by the agent.

code-quality