yc-reader — Y Combinator Company and Batch Data Query Skill

Skill Overview

yc-reader is a read-only Y Combinator data query skill that retrieves startup ecosystem data such as YC company profiles, batch lists, industry and tag distributions, and hiring status through the open-source yc-oss API. No API key is required; filtering can be performed using curl together with jq.

Use Cases

1. Sector and investment research: Retrieve lists of YC companies by industry or tag—for example, identify YC startups in fintech, healthcare, or AI, and combine team size and status fields for preliminary quantitative analysis to support startup research reports or background checks during investment due diligence.

2. Job search and hiring tracking: Query YC companies that are currently hiring (approximately 1,400 companies), or focus on a specific batch, such as W25 or S24, to view new company lists and quickly identify teams showing signs of growth.

3. Startup ecosystem and trend analysis: Retrieve aggregate data such as the total number of companies, batch lists, and industry and tag catalogs; observe changes in sector distribution across batches from different years, or compare the composition of top YC companies.

Core Features

1. Company and batch queries: Supports retrieving predefined lists of all companies, top companies, hiring companies, nonprofits, and more. Batch queries use the {season}-{year} naming convention, such as winter-2026 or fall-2025, and can be traced back as far as summer-2005. Individual company profiles can also be retrieved through batches/{batch}/{slug}.json, such as batches/summer-2009/stripe.json.

2. Industry, tag, and diversity filters: Retrieve companies by industry (industries/fintech.json) or tag (tags/developer-tools.json). Additional endpoints provide diversity data, such as companies founded by Black founders, Latino founders, or women founders.

3. Client-side aggregation with jq: Dimensions not provided by the API can be filtered locally with jq—for example, filtering by isHiring, status, or team_size, extracting concise results by field, or sorting with sort_by(-.team_size) to retrieve the top N companies. This supports a wide range of custom analysis requirements.

Data source and limitations: The data comes from yc-oss/api, a third-party unofficial project whose source is YC’s Algolia search index. It is updated daily through GitHub Actions, so the data is close to real time. This skill is read-only and performs no write operations. Its index is limited to publicly launched YC companies; companies that have not been made public, or that have shut down without a preserved record, may not be searchable. companies/all.json contains approximately 5,700 companies and produces a relatively large response, so more precise endpoints such as batch, industry, or tag endpoints are recommended.

Frequently Asked Questions

Does yc-reader require an API key or login?

No. The yc-oss API is a free, public static JSON dataset that requires neither authentication nor registration; you can make requests directly with curl. Locally, only the curl and jq commands are needed, both of which are preinstalled on most systems. If jq is missing, macOS users can run brew install jq, while Debian/Ubuntu users can run sudo apt-get install jq.

Is the data up to date? How often is it updated?

The data is updated once daily through GitHub Actions, so its status is close to real time. Fields that change frequently, such as hiring status and team size, should be fetched again when needed rather than relying on cached results.

Can it find all YC companies, including private ones?

No. The index covers publicly launched YC companies, approximately 5,700 in total. Companies that have not been made public, or that have ceased operations without being preserved in the index, cannot be found. Therefore, statistical results should be understood as representing a public sample rather than YC’s complete investment portfolio.

Can yc-reader modify or write data?

No. This is a purely read-only data source. The API provides only static JSON files and has no write endpoints. It is intended for queries and research and cannot be used to submit or update company information.

How do I find all companies in a particular batch?

Batch names use the lowercase, hyphenated {season}-{year} format. For example, to query the Winter 2025 batch:

curl -s https://yc-oss.github.io/api/batches/winter-2025.json | jq length

To view the companies in that batch that are currently hiring, add a filter:

curl -s https://yc-oss.github.io/api/batches/winter-2025.json | jq '[.[] | select(.isHiring == true)]'

If you are unsure of the batch name, first retrieve all valid names from batches[].name in meta.json.

How do I find YC companies that are hiring?

Request the dedicated hiring endpoint directly:

curl -s https://yc-oss.github.io/api/companies/hiring.json | jq '.[] | {name, one_liner, batch, team_size, website}'

Approximately 1,400 companies are listed. You can also combine this with an industry endpoint for further filtering, such as viewing only hiring companies in fintech.

What is the data source? How accurate is it?

The original data source is YC’s official Algolia search index, which is fetched and organized into static JSON files by the open-source yc-oss community project and updated daily. The field contents match YC’s public pages, but this is an unofficial project. If authoritative information or guarantees of field-level accuracy are required, verify the data against YC’s official pages.

Is querying free? Are there usage limits?

It is completely free, with no request quotas or rate limits—the API provides static files hosted on GitHub Pages, and each request simply retrieves a JSON file. The only point to note is that companies/all.json is relatively large; frequent downloads may be slow, so use more specific endpoints whenever possible.