CratesYour Personal Internet Search

Boily

Boily is a South Korea-focused service that measures how often hospitals appear in recommendations from ChatGPT, Claude, Gemini, and Perplexity. It runs recurring tests every two weeks, compares results with nearby competitors, and breaks visibility down by question and AI engine. Hospitals can also hire Boily for GEO and AEO work, including rewriting text-heavy website content into formats that search crawlers and generative systems can read more reliably. The service does not promise first-place rankings or guaranteed exposure. Instead, it offers a baseline, ongoing reports, and optional website optimization for clinics that want to understand how AI assistants may influence local healthcare discovery.

Japanly AEO

Japanly AEO is an independent visibility checker for brands targeting Japanese AI search. It tests how often a brand appears in Japanese answers from ChatGPT, Claude, and Gemini, while also examining honorific language and whether site markup is easy for AI systems to cite. The tool returns a red, yellow, or green assessment, example prompts, and a prioritized repair checklist. The first scan is free, takes about two minutes, and does not require a credit card. It is aimed at overseas SaaS companies, B2B marketers, and SEO teams that need to understand whether their brand is being surfaced in Japanese AI recommendations.

SerpApi

SerpApi turns search engine results into structured JSON that developers can use in applications, automation, and AI agent workflows. The platform covers Google, Bing, Amazon, YouTube, social networks, shopping sites, and more than 100 other search and vertical-search sources. It handles much of the operational work behind search access, including proxy rotation, CAPTCHA challenges, and result parsing. Developers can connect it through conventional APIs, SDKs, MCP servers, and Skills-oriented integrations. A monthly allowance of 250 free searches makes it practical for prototypes, although production projects need to account for usage-based costs and the learning curve of working with different response schemas.

Atlas

Atlas is listed at atlasppt.com but the homepage did not return a public product page at check time, so features and pricing cannot be independently confirmed here.

Sylgeo

Generative engine optimization platform that tracks brand visibility across ChatGPT, Claude, Gemini, and Perplexity and monitors competitor mentions.

Trakkr Data

Trakkr Data opens up AI search telemetry as live, explorable datasets, free for anyone. It includes eight datasets covering top AI-recommended brands, cited sources, crawler fetches, model comparisons, query rewriting, and web adoption. Updated daily with an API. Their data shows models agree on recommended brands only about 43% of the time.

gpt-researcher: Open-Source Deep Research Agent

GPT Researcher is an open-source deep research agent for web and local research, producing cited reports on any task. It aggregates 20+ sources, generates 2,000+ word reports, and cuts hallucination. Built with Python, licensed under Apache-2.0. As of collection time, it has 27,864 stars.

Vespa: The Open-Source AI Search Platform

Vespa is an open-source AI search platform built on Java, boasting over 7,000 GitHub stars and nearly 100,000 commits. It positions itself as the core search infrastructure for the AI era, ideal for teams needing to build their own robust search services. This overview introduces its current status and target audience.

XERJ: AI Data Indexing Beyond grep

XERJ is an open-source data indexing engine written in Rust for AI search workflows. It automatically indexes code, documentation, logs, and PDFs so agents can query relevant content instead of repeatedly scanning files with grep or sed. The project claims to use 40 times fewer tokens than grep, although that figure has not been independently validated or explained with detailed test conditions. XERJ also advertises Elasticsearch compatibility, which could make it easier for teams to reuse existing clients and query patterns. It may fit RAG pipelines, security audits, code search, and persistent agent memory, but its documentation and production-readiness details are still limited.

AutoClip: AI Video Toolkit for Downloading and Highlight Editing

AutoClip is an AI video toolkit that downloads video sources from YouTube and Bilibili, detects highlights using Qwen models, and exports edited collections. The project primarily uses Python and React, and is licensed under MIT.