daehwan kim

Scrape any webpage with a natural-language prompt instead of CSS selectors: describe what you want and ScrapeGraphAI's LLM extracts it as structured JSON. SmartScraper, Markdownify and SearchScraper modes. Wraps the open-source ScrapeGraphAI project (MIT).

Read invoices, receipts and bills automatically and get structured data back: vendor, line items, totals, tax, dates and currency. Accepts PDF and image files, returns clean JSON, and runs as an MCP server so an AI agent can call it directly.

AI code review for pull-request diffs: severity-rated comments on security, performance and maintainability, plus plain-English explanations of complex code. Supports Python, JavaScript/TypeScript, Go, Java and Rust, and runs as an MCP server an agent can call.

Turn a list of domains into complete company records: name, description, logo, tech stack, social profiles, public contacts and WHOIS registration. Built for lead enrichment and CRM cleanup. Domains that return nothing are never charged.

Rewrite, translate, summarize and repurpose content at scale: blog to social, long to short, formal to casual, across 40+ language pairs. Feed batches of text in one run and get clean structured output. Runs as an MCP server an AI agent can call.

Scrape Coupang product reviews and get a computed verdict per product: rating distribution, top complaint keywords, per-option rating gaps and a one-line headline. Korean-native. Unofficial tool, not affiliated with Coupang.

Convert data between CSV, JSON, XML, Excel, YAML and TSV at scale — no manual reformatting. Control delimiters and encoding, validate against a schema, and run whole batches of files in one pass. Built for ETL jobs, data migration and developer pipelines.

Score any open-source GitHub repository before you depend on it: security posture, maintainership health, dependency risk and license compliance, combined into a single rating. Built for OSS due diligence in enterprise adoption reviews and VC technical diligence.

Convert PDF documents into clean, readable Markdown with the document structure preserved — headings, lists and paragraph flow. Process files in bulk and get output that drops straight into LLM pipelines, RAG systems, knowledge bases and documentation sites.

Analyze vehicle damage from photos with AI vision: detect dents, scratches and broken parts, then return a severity score and a repair cost range. Batch a whole claim at once — built for insurers, rental fleets and used-car platforms.

Bulk YouTube transcript and subtitles extractor. Up to 200 video URLs, a channel or a playlist in; one row per video out: full text, timestamped segments, language, manual or auto-generated, word count, duration, verdict. Multilingual with auto-translation. No API key, no cookies, no video download.