⚡ YukisDB — Unified Developer Cloud BaaS
Sub-millisecond NoSQL JSON Database • 2GB Object Storage (Zero Egress Bills) • AI Vector Memory • Native MCP 2024-11-05
🌐 Web Portal • 📖 Documentation • 📜 OpenAPI 3.1 Spec • 🤖 Agent LLMs.txt • 💬 Support
🌟 Why YukisDB?
YukisDB (hosted on yukiapi.site by SUDEEPBOTS) is an all-in-one developer cloud backend engine built from the ground up for modern full-stack web apps, autonomous AI agents, and developer toolchains:
- 🚀 Sub-millisecond NoSQL JSON Store: Ultra-fast document CRUD, flexible filtering, aggregation pipelines, and automatic schema indexing.
- 📦 2GB Object Storage with Zero Egress Billing: Upload datasets, videos, model weights, and media files up to 2 GB per file with instant HTTP 206 partial streaming and $0 bandwidth bills.
- 🧠 AI Vector Memory & Semantic Search: Native nearest-neighbor cosine similarity search for long-term agent memory and RAG context pipelines.
- 🤖 Model Context Protocol (MCP 2024-11-05): Direct Streamable HTTP & SSE transport for Claude Desktop, Cursor AI, Windsurf, and AutoGPT.
- 🧪 Zero-Friction Sandbox Mode: Test persistence and queries at
/api/v1/sandbox/insertwithout mandatory account registration or credit cards. - 🔒 Enterprise-Grade Security: SHA-256 one-way hashed API keys, CIDR IP whitelist security, TLS 1.3 in-flight encryption, and AES-256 disk encryption at rest.
📦 Installation
Single command gives you both the Python SDK (import yukisdb) and the Developer CLI (yukisdb command):
pip install yukisdb
💻 Developer CLI Quickstart
Run yukisdb directly in your terminal:
# 1. Check Singapore Cluster Health & Telemetry
$ yukisdb health
# 2. View Developer Free Tier Quotas & Specs
$ yukisdb tier
# 3. Test Zero-Auth Sandbox Document Insert
$ yukisdb sandbox insert '{"task": "Evaluate YukisDB", "status": "completed"}'
# 4. Query Sandbox Documents
$ yukisdb sandbox find '{"status": "completed"}'
# 5. Generate Instant Ephemeral API Key for AI Agents
$ yukisdb anonymous-key
# 6. Authenticated Database Operations (Export your API Key)
$ export YUKISDB_API_KEY="ydb_live_your_key_here"
# Insert Document
$ yukisdb insert users '{"name": "Sudeep", "role": "admin", "skills": ["python", "ai"]}'
# Find Documents with Query
$ yukisdb find users '{"role": "admin"}'
# Upload 2GB File to Object Storage
$ yukisdb upload model_weights.pt custom-slug-2026
# List Stored Files
$ yukisdb list-files
# Start Local MCP stdio Bridge for Claude Desktop / Cursor
$ yukisdb mcp
🐍 Python SDK Guide
1. Initialize Client
from yukisdb import YukisDB
# Automatically reads YUKISDB_API_KEY from environment if not passed explicitly
db = YukisDB(api_key="ydb_live_your_api_key_here")
2. NoSQL Document Persistence (CRUD)
# Insert a new document
user = db.collection("users").insert({
"name": "Alice Johnson",
"email": "alice@example.com",
"tier": "developer",
"active": True
})
doc_id = user["id"]
print(f"✅ Document created with ID: {doc_id}")
# Find documents with JSON filter
active_devs = db.collection("users").find({"tier": "developer", "active": True})
print("Found developers:", active_devs)
# Get document by ID
doc = db.collection("users").get(doc_id)
# Update document fields
db.collection("users").update(doc_id, {"tier": "enterprise"})
# Run Aggregation Pipeline ($match, $group, $sort)
stats = db.collection("users").aggregate([
{"$match": {"active": True}},
{"$group": {"_id": "$tier", "total": {"$sum": 1}}}
])
# Delete document
db.collection("users").delete(doc_id)
3. AI Vector Embeddings Search (Cosine Similarity)
# Query semantic nearest neighbors for agent memory
results = db.collection("agent_memory").vector_search(
vector=[0.124, -0.451, 0.882, 0.051],
field="embedding",
top_k=5
)
for item in results:
print(f"Match: {item['title']} | Similarity Score: {item['_score']}")
4. 2GB Object Storage (Zero Egress Bandwidth Bills)
# Upload any binary file up to 2GB with high-speed CDN URL
file_info = db.storage.upload(
file_path_or_bytes="4k_dataset_video.mp4",
custom_slug="dataset-singapore-2026"
)
print(f"Uploaded! Public Streaming CDN URL: {file_info['url']}")
# List all objects in storage
objects = db.storage.list()
print("Tenant files:", len(objects))
5. Zero-Auth Sandbox (Instant Prototyping)
# No API key needed for sandbox operations
sandbox_doc = db.sandbox.insert({"experiment": "LLM agent autonomous test"})
print("Persisted into sandbox:", sandbox_doc)
🤖 Model Context Protocol (MCP 2024-11-05) Setup
YukisDB provides first-class support for Model Context Protocol (MCP). Connect Claude Desktop, Cursor AI, Windsurf, or AutoGPT directly:
Claude Desktop Configuration (claude_desktop_config.json):
{
"mcpServers": {
"yukisdb": {
"command": "yukisdb",
"args": ["mcp"],
"env": {
"YUKISDB_API_KEY": "ydb_live_your_key_here"
}
}
}
}
Direct Streamable HTTP & SSE Transport:
- Streamable HTTP Endpoint:
https://yukisdb.yukiapi.site/mcp - SSE Stream Endpoint:
https://yukisdb.yukiapi.site/mcp/sse - MCP Manifest:
https://yukisdb.yukiapi.site/.well-known/mcp/manifest.json
Supported MCP Tools:
insert_document: Save JSON document memory.find_documents: Query documents with JSON filters.get_document: Retrieve document by ID.update_document: Update document attributes.delete_document: Delete document by ID.aggregate_documents: Multi-stage aggregation pipelines.vector_search: Cosine similarity nearest-neighbor semantic search.upload_file: Upload files & datasets (up to 2GB).list_files: List tenant storage files.get_cluster_health: Real-time datacenter status.get_tier_policy: Free tier quotas & parameters.
⚙️ Architecture & Reliability
| Specification | Value |
|---|---|
| Datacenter Region | Singapore (AP-Southeast), Tier-1 Redundant Network |
| Document Query Latency | Sub-millisecond indexed NVMe SSD Lookups |
| Object File Size Limit | Up to 2 GB per file (Unlimited total files) |
| Bandwidth Policy | Free Unlimited Egress for Developer Accounts |
| Encryption In-Flight | TLS 1.3 Strict HTTPS Transport |
| Encryption At-Rest | AES-256 Disk Encryption |
| Key Security | SHA-256 One-Way Hashing with Salt |
| Rate Limit Standard | RFC 6585 / IETF RateLimit Headers (RateLimit-Limit: 1200) |
| Error Handling Model | RFC 9457 Problem Details (application/problem+json) |
| API Versioning | Explicit /api/v1/... with RFC 8594 Sunset Guarantees |
🔗 Developer Resources Index (yukiapi.site)
- 🌐 Portal & Dashboard: https://yukisdb.yukiapi.site
- 📖 Developer Documentation: https://yukisdb.yukiapi.site/docs
- 🔐 Authentication & Scopes: https://yukisdb.yukiapi.site/docs/auth
- ⚡ Webhooks & Events: https://yukisdb.yukiapi.site/docs/webhooks
- 📜 OpenAPI 3.1 Spec (JSON): https://yukisdb.yukiapi.site/openapi.json
- 📄 OpenAPI 3.1 Spec (YAML): https://yukisdb.yukiapi.site/openapi.yaml
- 🤖 LLMs Reference (llms.txt): https://yukisdb.yukiapi.site/llms.txt
- 📚 Full Reference (llms-full.txt): https://yukisdb.yukiapi.site/llms-full.txt
- ⏳ Deprecation & Versioning Policy: https://yukisdb.yukiapi.site/deprecation
- 🗺️ Sitemap: https://yukisdb.yukiapi.site/sitemap.xml
- 🐙 GitHub Repository: https://github.com/SUDEEPBOTS/yukisdb
📄 License
YukisDB is licensed under the MIT License.
Copyright © 2026 SUDEEPBOTS. All rights reserved.
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file yukisdb-1.0.1.tar.gz.
File metadata
- Download URL: yukisdb-1.0.1.tar.gz
- Upload date:
- Size: 12.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
901ee75aca30ffac160145409d16c4264aba2a3a3e22ffd88d67f55cc7f18d33
|
|
| MD5 |
5b209845f04e8c6b3e710c3bbbcaf1d7
|
|
| BLAKE2b-256 |
50871dbbbbdc3e4543daa4f98891fd7609bfe9b096ac6a2224823509121a98ae
|
File details
Details for the file yukisdb-1.0.1-py3-none-any.whl.
File metadata
- Download URL: yukisdb-1.0.1-py3-none-any.whl
- Upload date:
- Size: 10.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ca4463056b3f24c355ecdaf7ad1cc967220edb2792f05ed89ea5a81b8ba0762d
|
|
| MD5 |
3156facccc9223b7456741a3d77ea1ed
|
|
| BLAKE2b-256 |
980a364ae36697873956711f735c5d766508c5dfedde5925cdeba93dca8eded4
|