SearchCandy Labs

Active

Tagline:Document retrieval for AI agents

Updated:October 22, 2026

Domain Rating:

Social Media:

AI Knowledge Base AI Agent AI Search Engine AI Document Extraction AI Developer Tools API Available Web App Cloud-based For Developers AI Agent Ready

Open Website

Visit Website

What is SearchCandy Labs?

Temporal Graph Memory (TGM) turns your documents into a knowledge graph and answers questions with exact passages from them. There is no generative language model in the retrieval engine — returned passages come from your documents. Every change to a graph is saved as a version, so answers stay auditable after your documents change.

How to use SearchCandy Labs?

1. Book a 30-minute call to set up pilot access. 2. Receive a password-free sign-in link by email and open the console. 3. On first sign-in, copy your API key (shown exactly once) and set SEARCHCANDY_API_KEY in your environment. 4. Send documents through the API and ask questions in plain language to get cited passages back.

SearchCandy Labs Core Features

Cited passages, not guesses

TGM returns the exact passage from your documents that answers a question, naming the source file. If the answer is not in your documents, it returns not_covered: true instead of inventing one.

Temporal Graph Memory

Documents become a knowledge graph; passages are connected using embeddings alone, so related context is retrieved together rather than in isolation.

Versioned and auditable

Every change to a graph is saved as a version. Ask last month's version a question and get last month's exact answer, keeping results checkable over time.

No generative LLM in retrieval

The retrieval engine uses embeddings only. Returned passages come from your documents, so every answer is grounded in a source you can inspect.

SearchCandy Labs Use Cases

Ground an AI agent in your own documents with citations it can show

Keep auditable answers for teams that must trace responses to a source

Retrieve a precise passage across a document set by asking in plain language

Build retrieval-augmented workflows without a generative model in the loop

SearchCandy Labs Frequently Asked Questions

How does Temporal Graph Memory work?

You send documents through the API. TGM splits them into passages and connects them into a graph using embeddings alone. You ask a question in plain language and get the exact passages back, each naming its source file.

What happens if the answer isn't in my documents?

You get not_covered: true instead of a guess. The retrieval engine never invents an answer; it only returns passages that exist in your files.

How do I get an API key?

Access is set up for you during a pilot. Book a call, then after onboarding you receive a sign-in link. On first sign-in the console shows your API key exactly once; save it, because it is not stored and cannot be shown again.

Are my answers reproducible over time?

Yes. Every change to a graph is saved as a version. Asking the same question against a past version returns that version's exact answer, so results stay auditable after your documents change.

SearchCandy Labs Product Background & Official Links

SearchCandy Labs Support Email & Customer service contact & Refund contact etc.

SearchCandy Labs Support Email: hello@searchcandy-labs.com

More Contact, visit the about us page(https://www.searchcandy-labs.com/docs)

SearchCandy Labs Company Info

SearchCandy Labs Company name: SearchCandy Labs Inc.

SearchCandy Labs Developer Type: company

More about SearchCandy Labs, Please visit the about us page(https://www.searchcandy-labs.com/docs)

SearchCandy Labs Pricing & Plans

Pilot

contact

Contact for pilot access

30-minute onboarding call to set up your account
API key and SDK access
Versioned graphs with full answer history
No per-seat signup form required

Subscribe & upgrade now

SearchCandy Labs Tags

API Available Web App Cloud-based For Developers AI Agent Ready

Best Suited Occupations for SearchCandy Labs

Software Developers Data Scientists Academic Researchers