querying-json

The querying-json skill leverages the power of jq to perform surgical data extraction from structured JSON documents. It is designed for software engineers, data analysts, and developers who need to interact with large configuration files, massive API responses, or complex dependency manifests without bloating the LLM context window. By targeting only the necessary data points—such as version numbers, specific dependency trees, or configuration keys—users can maintain higher model performance and lower latency in their workflows.

• Precise field extraction using jq query syntax for nested objects, arrays, and conditional filtering.

• Significant context optimization, saving 80-95% of token usage by avoiding full file reads for large JSON data.

• Seamless pipeline integration with other CLI tools like fzf for interactive selection, sd for transformation, and bat for formatted previewing.

• Support for advanced data manipulation, including array indexing, key extraction, object construction, and default value handling.

• Ideal for managing project infrastructure, such as parsing package.json, tsconfig.json, lock files, and automated deployment configurations.

• Use this skill when dealing with files larger than 50 lines where only a subset of information is required for the current task.

• For small files or scenarios where the overall structure must be understood for architectural refactoring, direct file reading remains the preferred alternative.

• Always favor this skill over executing raw bash commands to ensure consistent, secure, and context-efficient data retrieval.

• Compatible with common Unix/Linux pipelines; use jq to transform JSON into flat text, CSV, or other formats before piping into further utilities.

• For complex queries involving data transformation patterns, error handling, or array manipulation, refer to the integrated jq-guide documentation provided within the skill directory.