Connect your agent
to the real world.
SandBase makes real-world capabilities as reliable and priceable as APIs — find them, build an agent, and deliver the work.
Tell your agent.
One installer, guided authorization in the browser, and the capabilities you picked show up in the AI tools you already use.
$ curl -fsSL https://sandbase.ai/install.sh | sh -s -- --client codex
→ guided authorization — no keys land in your shell historyOne platform. Three ways in.
Use it inside the AI tools you already have, call one capability from your product, or deliver a full agent. One key covers all three.
Setup
Use in your local AI tools
Pick capabilities from the Store, run Setup once, and use them in Codex, Claude Code.
Start with SetupAPI
Call one API from your product
Copy a working request for a model, search, data, or SaaS API — the price is on the card.
Browse APIsAgent
Build a reusable agent
Define the job, give it tools, test the result, then publish it as an endpoint or a schedule.
Build an agent
What’s in the Store.
1,195 models and 1,105 real-world APIs — every card shows its price next to a working request.
- AGENTSEO Keyword InsightsSandBaseDiscovers, validates, and prioritizes keywords with DataForSEO.Runson usage
- LLMdeepseek/deepseek-v4-flashDeepSeekEfficiency-optimized MoE with a 1M-token context window.$0.14 in · $0.28 out/ 1M
- IMAGEgoogle/nano-banana-2/editGoogleState-of-the-art image generation and editing.$0.06/ call
- APIexa/searchExaNeural web search with page contents returned inline.$0.007/ call
- LLMClaude Opus 4.6AnthropicReasoning, coding, and tool use for long-running workflows.$5 in · $25 out/ 1M
- AGENTYouTube TranscriptSandBaseExtract transcripts and captions for research and summaries.Free to clone
- APIInstagramInstagramProfiles, posts, reels, and comments for creator research.View pricing/ endpoint
- SKILLYouTube ResearchSandBaseResearch videos, channels, and audiences from your AI tools.Public Skill
Every run
leaves a record.
Each execution becomes a run with its input, output, trace, and status — and Activities shows what it cost.
- Input & output
- The request that started it. The result it delivered.
- Trace & status
- Every step along the way, kept with the run.
- Usage & cost
- Activities shows what each execution cost.
Deliver it, then operate it.
Test before you publish. Deliver as an endpoint or a schedule, get webhooks when work finishes or fails, and see cost per run.
Authentication
One workspace key — bearer or x-api-key across every capability.
Testing
Run the agent and see the result before anything depends on it.
Publishing
A published agent is an API your apps can call.
Schedules
Recurring runs without keeping anything alive.
Webhooks
Notifications when async work finishes or fails.
Activities
Request history and usage per call, run, and schedule.
Start with a run, not a plan.
Free credits
$1
Try models, APIs, Setup, and agent runs before you scale.
Start free ›Pay as you go
$0
Prices live on each model and API card; agent runs pay for the work performed.
Create an API key ›Team & enterprise
Custom
Volume usage, SSO, invoice billing, and private deployment options.
Contact sales ›