Se

Searchable APIs for your AI Agents

Hacker News

Searchable APIs for your AI Agents

Going live in 12 hours. Quick run-down of it: - Most software, platform, & infrastructure vendors already provide OpenAPI specifications for all read & write operations on their services. - On giving LLMs access to those services, the usual approach is to provide hard-coded tools. - Problem # 1: Redundancy. When you hard-code tools, you end up creating wrappers, unnecessary layers, and you're forced to include/exclude parts of the spec, you're forced to update your wrapper layer when they update their specs. - Problem # 2: They can get stuck on edge cases that requires a tool that isn't provided before-hand. They need to be able to adapt their capabilities based on the problem at hand. Like trinity in matrix downloading a skill to fly a helicopter - in real-time. - Problem # 3: Trying to fit entire API specs in their context windows isn't possible - and it doesn't make sense too. It's like requiring a human to read an entire book when he just needs few pages from it to answer the problem at hand. - To solve it, our thesis is that as long as LLMs can execute HTTP requests, they should be able to communicate with all SaaS, PaaS, & IaaS vendors. - The key ingredient is to create a search engine of these OpenAPI Specifications. - This 1.) removed the redundancy wrapper layer, 2.) lets them adapt to any situation they're in, and 3.) saves tokens by not requiring entire specs to be in their context window. In the screenshot below, I told the agent "List my digitalocean servers, my api key is dop_v1_91ece4311ff2582ebb55889716c992fa247462729ff3a0bdc5d400dd75c302d2." Here's what happened: - It searched for "list digitalocean droplets" - It searched for "list digitalocean domains" - It retrieved the operation schema for listing droplets at /api/operations/266 - It retrieved the operation schema for listing domains at /api/operations/256 - It retrieved the droplets at http://api.digitalocean.com/v2/droplets - It retrieved the domains http://api.digitalocean.com/v2/domains - It responded to the user --- RESPONSE START --- I've retrieved information about your DigitalOcean account. Here's what I found: SERVERS (DROPLETS): 1. ubuntu-c-2-sfo3-01 - ID: 472556983 - Status: active - Region: San Francisco 3 (sfo3) - Size: c-2 (2 vCPUs, 4GB RAM, 25GB disk) - OS: Ubuntu 24.04 (LTS) x64 - IP Addresses: - Public: 64.23.132.206, 24.199.78.229 - Private: 10.124.0.2 2. ubuntu-s-1vcpu-2gb-sgp1-01 - ID: 483339741 - Status: active - Region: Singapore 1 (sgp1) - Size: s-1vcpu-2gb (1 vCPU, 2GB RAM, 50GB disk) - OS: Ubuntu 24.10 x64 - IP Addresses: - Public: 159.223.56.197 - Private: 10.104.0.3 DOMAINS: You don't have any domains registered in your DigitalOcean account. Would you like me to perform any specific actions with these resources or would you like more detailed information about any of these droplets? --- RESPONSE END ---

Share card

Actual performance

2points
Did not reach leaderboard

Launch Intel predictions

Analyze your own launch →
Indie HackersFits the IH revenue-focused audience · Missing: supports, reddit linkedin, podcasting
84%84% predicted probability of success on Indie Hackers, based on ML models trained on real launch data.
best fitHighest predicted score across all platforms for this description.
Product HuntOn track for Day 1 leaderboard · Strong signals: agents, agent, user · Missing: mac, macos, cursor
84%84% predicted probability of success on Product Hunt, based on ML models trained on real launch data.
Hacker NewsStrong engagement from HN community · Strong signals: ide, io · Missing: https docs, excited, just released
53%53% predicted probability of success on Hacker News, based on ML models trained on real launch data.
nativeThis product was originally launched on this platform.
TrustMRRLess likely to generate early MRR · Missing: mobile apps, ios, personal
36%36% predicted probability of success on TrustMRR, based on ML models trained on real launch data.
AppSumoMay struggle as an AppSumo deal · Strong signals: platform · Missing: plus, intuitive, reviews
33%33% predicted probability of success on AppSumo, based on ML models trained on real launch data.
Acquire.comPre-revenue stage for this audience · Strong signals: saas, active · Missing: arr, mrr, revenue
16%16% predicted probability of success on Acquire.com, based on ML models trained on real launch data.
BetaListMay not resonate with beta-testers · Missing: web3, chat, crypto
0%0% predicted probability of success on BetaList, based on ML models trained on real launch data.

Incorrect prediction on native model

Similar products

Pa
Parlel – LinkedIn, but searchable by AI agents33%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Parlel – LinkedIn, but searchable by AI agents

Hacker News3
Cl
Clark Hash, 32x smaller searchable sketches for embeddings68%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Clark Hash, 32x smaller searchable sketches for embeddings

Hacker News1
Co
ConGrep – Searchable Congress52%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

ConGrep – Searchable Congress

Hacker News2
Fi
Fighting ICO Scammers with APIs51%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Fighting ICO Scammers with APIs

Hacker News9
Op
OpenAPI OAuth 2.0 scopes enforcement for APIs66%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

OpenAPI OAuth 2.0 scopes enforcement for APIs

Hacker News7
Ko
Kompy, a Wrapper for Komoot APIs30%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Kompy, a Wrapper for Komoot APIs

Hacker News1
AI
AI Agents Benchmarking and Competition41%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

AI Agents Benchmarking and Competition

Hacker News2
OA
OAuth for AI Agents37%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

OAuth for AI Agents

Hacker News8
47
47jobs – A Fiverr/Upwork for AI Agents43%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

47jobs – A Fiverr/Upwork for AI Agents

Hacker News20
SA
SAIA – SCUMM for AI Agents25%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

SAIA – SCUMM for AI Agents

Hacker News3