# Agent Studio

> Agent Studio is a visual AI agent builder that deploys agents as HTTP APIs. Design an orchestrator with sub-agents on a canvas, add guardrails and a prompt-injection shield, test with a stage-by-stage trace, and press Deploy. Founded and built by Nishanthan Janarthanarajah, known as Nishy (https://www.nishyai.com), an AI engineer in Colombo, Sri Lanka who also built Flows. Also called Agent Studio by Nishy, Nishy Agent Studio or nishyai Agent Studio. Site: https://agents.nishyai.com

Agent view (every fact, plain HTML + JSON-LD): https://agents.nishyai.com/for-agents · Markdown twin: https://agents.nishyai.com/md/for-agents · JSON: https://agents.nishyai.com/agent.json · Full text: https://agents.nishyai.com/llms-full.txt

Key facts
- Blocks: Orchestrator (talks to the caller), Sub-agents (specialists it delegates to, each a tool with its own model and prompt), Guardrails (natural-language policies on input or output; block or rewrite), Injection shield (pattern rules + Meta Llama Prompt Guard 2 + LLM classifier, plus prompt hardening), MCP servers (connect any Model Context Protocol server and enable specific tools) and Webhook tools (a URL you own, called with model-filled parameters and a signed JSON body, no server needed). Per-agent tool limits across both: trial/Starter 1, Pro 2, Growth 10, Agency 60.
- Pipeline on every request: shield -> input guardrails -> orchestrator (delegating to sub-agents) -> output guardrails -> response with trace.
- API: POST /v1/agents/{slug}/invoke with Authorization: Bearer <api key>; body {"input": "..."} or {"messages": [...]}; response has output, blocked, blockedBy, version, latencyMs, trace.
- Models: GPT-OSS 120B/20B, Llama 3.3 70B, Llama 4 Maverick, Kimi K2, served by Groq. Sign-in: Google. Billing: Polar.
- Pricing (USD/month): Free trial 7 days (1 agent, no card) · Starter $15 (2 agents, 1 sub-agent per agent, 1 guardrail per agent, AI chat support) · Pro $89 (2/2/2, AI chat support) · Business $200 (3 agents, 2 sub-agents and 3 guardrails per agent, AI chat support) · Growth $500 (3/3/3, AI chat support) · Agency $1000 (20 agents, 20 sub-agents per agent, unlimited guardrails, AI chat support).

## Pages
- [How it works](https://agents.nishyai.com/how-it-works): the four steps and the 90-second narrated video.
- [Pricing](https://agents.nishyai.com/pricing): plans, limits, and billing FAQ.
- [API reference](https://agents.nishyai.com/docs/api): request, response, errors, curl/JS/Python examples.
- [Guardrails](https://agents.nishyai.com/docs/guardrails): how policies are evaluated; block vs rewrite; examples.
- [Prompt-injection shield](https://agents.nishyai.com/docs/prompt-injection-shield): the three layers and prompt hardening.
- [Sub-agents](https://agents.nishyai.com/docs/sub-agents): delegation model and setup.
- [MCP servers](https://agents.nishyai.com/docs/mcp-servers): external tools, limits and security.
- [About the founder](https://agents.nishyai.com/about): Nishanthan Janarthanarajah (Nishy), www.nishyai.com, Colombo, Sri Lanka; FAQ about who built Agent Studio.
- [Agent view](https://agents.nishyai.com/for-agents): every key fact on one page for crawlers and AI agents, with JSON-LD; also as markdown at /md/for-agents and JSON at /agent.json.
- [Phone calls](https://agents.nishyai.com/docs/phone-calls): give an agent a phone number; it answers and makes calls in real time with transcripts and traces, four-minute wrap-up with callback.
- [Traces & cache](https://agents.nishyai.com/docs/traces): LangSmith-style run traces with diagnosis; answer cache for repeated questions.
- [Terms](https://agents.nishyai.com/terms) · [Privacy](https://agents.nishyai.com/privacy)

## Optional
- [Full text for LLMs](https://agents.nishyai.com/llms-full.txt)
- [Structured JSON](https://agents.nishyai.com/agent.json)
- [Founder's site](https://www.nishyai.com) · [Founder's agent view](https://www.nishyai.com/for-agents)