How does it handle missing data in the PRD?

If a section or table is missing or empty, the skill returns an empty array ([]) for that category to maintain JSON schema consistency.

Can I use this skill with any Markdown file?

Yes, as long as the file follows the expected 'All Needed Context' heading structure and uses markdown tables for data, it can parse any .md file.

Does it support nested JSON objects?

The skill flattens table rows into a list of objects based on the specific schema defined for each context subsection (e.g., path, purpose, and priority for code files).

What specifically does this skill look for in a PRD?

The skill searches for a second-level heading titled '## All Needed Context' and parses specific subsections like Code Files, Docs, Examples, Gotchas, and External Systems.

Is this skill required for Flowspec commands?

Yes, it is a core component invoked by /flow:implement, /flow:generate-prp, and /flow:validate to gather necessary technical context.

Context Extractor

Name: Context Extractor
Author: jpoley

byjpoley

•

生产力与工作流

Transforms Product Requirements Documents (PRDs) into machine-readable JSON context for automated development workflows.

The Context Extractor skill bridges the gap between human-readable project specifications and automated execution by parsing Markdown-based PRDs. It specifically targets 'All Needed Context' sections to identify relevant code files, documentation, implementation examples, and external system dependencies. By converting these unstructured tables into structured JSON, it enables Claude and the Flowspec framework to intelligently prioritize file analysis, avoid known pitfalls, and validate technical requirements before implementation begins.

主要功能

01Data Sanitization: Strips markdown artifacts like backticks and extra whitespace for clean programmatic use.

02Priority Mapping: Captures 'Read Priority' levels to optimize how AI models consume codebase context.

03Markdown Table Parsing: Converts complex markdown tables into structured, machine-readable JSON objects.

04Error Diagnosis: Provides specific remediation steps for missing sections or malformed markdown syntax.

055 GitHub stars

06Categorized Extraction: Automatically organizes data into code files, documentation, examples, gotchas, and external systems.

使用场景

01Verifying that all prerequisite files and documentation links exist before starting a development sprint.

02Automating the preparation of context bundles for the /flow:generate-prp workflow.

03Providing structured implementation guidance to Claude during /flow:implement tasks.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jpoley/flowspec context-extractor

For use in Claude.ai and ChatGPT