AI Agent Stack for Tel Aviv Security and Research Builders
Tel Aviv is where Israeli security startups and university labs ship product. If you need an AI agent API Tel Aviv teams can share, SandBase is one key for models, data APIs, hosted sessions, and a sandbox.
What a Tel Aviv security team should not split apart
A product group here often already has a detection model. It still collects evidence, runs the agent, and executes the follow-up script in three places. SandBase is one place:
2,000+ models and data APIs across 100+ platforms, drawn from one balance
Agents you store and Sessions you launch, in the OpenAI Agents and OpenAgentCore shapes
An E2B-compatible sandbox for the files and commands that follow a finding
SKILL.md so a coding assistant can discover the catalog
Agents and sandboxes can begin with no charge. Eligible data APIs share 100 free calls per day per organization, resetting at 00:00 UTC. Model use is a separate bill—open the listing before a long investigation.
The base URL is https://api.sandbase.ai/v1. Send Authorization: Bearer $SANDBASE_API_KEY.
Tel Aviv in the public rankings
These pages describe the city. They are not a SandBase office.
Startup Genome’s Tel Aviv profile for the Global Startup Ecosystem Report 2026 ranks the city #4 globally, #1 in MENA, #5 globally for its AI-native cluster, and #1 for talent strength. It calls cybersecurity a defining pillar and notes that CyberWeek 2025 in Tel Aviv drew people from more than 90 countries.
The 5W AI City Index 2026 places Tel Aviv #4 among 25 metropolitan AI hubs, with a published score of 83.5.
Tel Aviv University says it leads Israeli institutions in the Times Higher Education World University Rankings 2027, and lists Cyber and AI & Data Science among its research areas.
SandBase is the API under that mix of security product and campus research: one key, not a new vendor for every sensor or model.
Tel Aviv workflows
1. A sourced brief before a security design review
Two builders in Tel Aviv need a cited note on a threat vertical: advisories, posts, and docs, duplicates removed. One agent, data APIs, and a Session replace a pile of vendor logins before the review.
2. An OpenAI Agents prototype for a security product
The prototype already uses the OpenAI Agents SDK. The team points it at https://api.sandbase.ai/v1, keeps client.beta.agents, and adds SandBase data tools plus a sandbox so a customer demo can show more than one source.
3. A week around Tel Aviv University
Students and early companies near that campus want an agent sandbox in Tel Aviv that starts without a new contract. Paste set up https://www.sandbase.ai/SKILL.md into the assistant, try the catalog, and keep the key when the experiment becomes a hosted Agent.
A first session from Tel Aviv
Create one SandBase API key in the Console for the team.
Register the Agent with a model, instructions, and tools via POST /v1/agents or the OpenAI SDK.
When projection_status equals ready, start a Session and send the task.
If the task needs files or a script, call the E2B SDK at https://sandbox.sandbase.ai.
Or connect SKILL.md and let the assistant discover, inspect, then run.
The documented OpenAI Python SDK pattern, with instructions for Tel Aviv security and research readers.
Python
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["SANDBASE_API_KEY"],
base_url="https://api.sandbase.ai/v1",
)
agent = client.beta.agents.create(
model="deepseek/deepseek-v4-flash",
name="Tel Aviv security brief",
instructions=(
"Write for a Tel Aviv security and research audience. "
"Compare at least two sources and cite each one."
),
metadata={"city": "tel-aviv", "team": "security"},
)
print(agent.id, agent.version) # agt_..., 1
# A Session waits until projection_status is ready.
Keep scripts and files in a sandbox. The Sandbox quickstart is the setup.
A security stack gathered onto one key
A security stack gathered onto one key
Piece
Often split in Israel
On SandBase
Models
A separate lab contract
One catalog behind one key
Threat and web data
Scrapers and extra logins
Data APIs on the same balance
The agent
A loop the team hosts
OpenAI Agents-compatible Agents and Sessions
Follow-up code
A runner under the team’s care
E2B-compatible sandboxes on the SandBase key
Coding assistants
MCP wired locally
SKILL.md for Cursor, Claude Code, Codex, and others
Looking for an AI agent API Tel Aviv security teams can adopt while keeping the OpenAI Agents SDK? SandBase is that API, with data tools and a sandbox on the same key.
Sources for the city
Link these public pages. Do not add a SandBase office, a phone number, or a latency claim for Tel Aviv:
The startup rankings, the cybersecurity pillar, and CyberWeek 2025 come from Startup Genome’s Tel Aviv page. The 5W release is the source for fourth place among 25 metros and the 83.5 score. Tel Aviv University’s English site is the source for its Israel ranking and for listing Cyber and AI and Data Science as research areas. SandBase does not claim a public Tel Aviv phone number or a city-specific latency SLA on this page.
FAQ
Is SandBase an AI agent API for Tel Aviv teams?
Yes. SandBase is a global API, and this page is the Tel Aviv guide. Teams use one key for models, data APIs, Agents and Sessions, and sandboxes. Being in Tel Aviv, or in Israel, is not required.
Does SandBase run a Tel Aviv office or publish an Israel phone line?
No. The product is API-first. This guide explains local security and research demand. It does not announce a public office or a local phone line.
Can a security product keep the OpenAI Agents SDK?
Yes. Set base_url="https://api.sandbase.ai/v1" and keep client.beta.agents. The key then also reaches SandBase models, data APIs, hosted OpenAgentCore, and E2B-compatible sandboxes.
Does every model or data source need its own key?
No. One SandBase key covers the models and APIs on the platform. Keep it on a server. Do not place it in a browser bundle.
How do charges and the daily data pool work?
Use draws on the SandBase balance, and each listing shows its price and unit. Agents and sandboxes can run with no charge. Eligible data APIs share 100 free calls per day per organization and reset at 00:00 UTC. Calls past the allowance bill at the listed rate. Model calls are separate.