canonical: https://jentic.com/apis/defispace.com/defispace # Defispace AgentsPodium account API (agent-facing subset) The AgentsPodium account API lets an agent create, configure, pay for, and watch hosted AI agent pods directly, without a human operator. It deploys a pod from a persona and a chosen engine, installs its model key, enables Hermes toolsets and connected tool servers, sets messenger channels such as Telegram, and manages the full lifecycle with pause, resume, rebuild, backup, and delete. It exposes catalogs of personas, engines, tools, models, and LLM providers, accepts plan payments in USDT, USDC, or Telegram Stars, and connects agents to each other through an A2A peer list and public directory. ## For AI agents Deploy, configure, pay for, and monitor hosted AI agent pods directly from an agent: set models, toolsets, MCP servers, and channels, manage the lifecycle, and connect agents over A2A across 52 endpoints. ## Scope Does not host general websites, databases, or non-agent workloads, and it does not author the agent's own logic. Use for deploying, configuring, and paying for hosted AI agent pods only. ## Capabilities - Deploy a new AI agent pod and read, upgrade, pause, resume, rebuild, or delete it - Configure an agent's model key, enabled toolsets and tool servers, messenger channels, and Telegram bot token - Browse the catalog of personas, engines, Hermes toolsets, curated LLM models, and LLM providers - Monitor a pod's liveness, RAM and disk usage, remaining term, and last build status - Take off-pod backups, list server-side snapshots, and export an agent's own data as a tarball - Pay for an agent's plan with USDT or USDC crypto deposits or Telegram Stars orders - Connect agents over A2A by setting peer lists, opting into the public directory, and browsing the catalog ## Use cases ### An Agent Provisioning Other Agent Pods Because the AgentsPodium account API is meant to be driven directly by an agent, an orchestrator connected through Jentic can deploy a new pod from a chosen persona and engine, install its model key, enable the toolsets it needs, and bring it online, all without a human operator. This turns agent provisioning into a programmatic step rather than a dashboard task. Example prompt: Create a new agent from a persona and engine, install its model key, enable a toolset, and confirm the pod is live ### Configuring a Hosted Agent AgentsPodium exposes each pod's configuration as discrete operations: install or clear the model key, set enabled toolsets and tool servers, choose which messengers the agent listens on, install a Telegram bot token, append owner instructions to the persona, and set the schedule timezone. An agent can reconfigure itself or a peer without redeploying from scratch. Example prompt: Set the enabled toolsets and tool servers on an agent, then install its Telegram bot token ### Lifecycle and Health Management Owners can manage a pod across its whole lifecycle: check liveness and current RAM and disk usage against the plan, find when the term ends, pause to stop billing resources, resume, rebuild from the last backup, and take an off-pod snapshot. This keeps a fleet of agent pods observable and recoverable from one API. Example prompt: Check a pod's liveness and usage, and if it is unhealthy, rebuild it from its last backup ### Paying for an Agent Plan AgentsPodium accepts plan payments without a card: open a USDT or USDC crypto deposit order for an agent's plan, or open a Telegram Stars order, then list the orders to confirm settlement. An agent can top up its own hosting term using the payment providers the platform supports. Example prompt: Open a USDT crypto-order for an agent's plan and then list its crypto orders to confirm the deposit ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/agents` | Create a new agent (deploy a pod) | | GET | `/agents/{id}` | Read one agent's full record | | PATCH | `/agents/{id}/tools` | Set the agent's enabled toolsets and tool servers | | POST | `/agents/{id}/pause` | Pause an agent to stop billing its resources | | POST | `/agents/{id}/rebuild` | Rebuild the pod from its last backup | | GET | `/agents/{id}/liveness` | Check whether the pod is up and answering | | POST | `/agents/{id}/crypto-order` | Open a USDT or USDC deposit order for a plan | | GET | `/personas` | List the public persona catalog | ## Key resources - **Agents** — Deploy, read, configure, upgrade, pause, resume, rebuild, back up, and delete agent pods - **Catalog** — List personas, engines, Hermes toolsets, curated LLM models, LLM providers, and tiers - **Payments** — Open and list USDT/USDC crypto orders and Telegram Stars orders, and activate paid subscriptions - **A2A Directory** — Set peer lists, opt agents into the public catalog, and browse listed agents - **Keys** — List, mint, and revoke account API keys and exchange an e-mail code for a token ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 48 / 100 - **Maturity:** Foundational - **Dimensions:** - Foundational Compliance: 59 / 100 - Developer Experience & Jentic Compatibility: 61 / 100 - AI-Readiness & Agent Experience: 25 / 100 - Agent Usability: 93 / 100 - Security: 60 / 100 - AI Discoverability: 75 / 100 - **View full report:** https://jentic.com/apis/defispace.com/defispace/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring AgentsPodium by hand means handling its bearer auth (an ak_live_ key or a session token from the e-mail verify flow), the persona, engine, tool, and model catalogs, and crypto or Telegram Stars payment orders. Through Jentic you install once, import AgentsPodium from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** AgentsPodium puts each pod's id in the URL path (`/agents/{id}/...`), so a rule can pin your agent to one pod and the operations you pick: it can read status, usage, and backups for that pod while deleting it, minting keys, or opening payment orders stay out of reach. You choose the operations it may call, so nothing destructive or billable runs unless you add it. - **Credential handling:** Your AgentsPodium API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'deploy an agent pod' or 'set an agent's toolsets', and Jentic returns the matching AgentsPodium operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Railway API** — General PaaS that deploys and manages app instances, versus AgentsPodium's purpose-built AI agent pods. - **Render API** — Managed cloud hosting for web services and workers, versus AgentsPodium's agent-specific lifecycle and A2A directory. - **E2B API** — Ephemeral cloud sandboxes for running AI-generated code, complementing AgentsPodium's persistent hosted pods. - **DigitalOcean API** — Raw cloud compute droplets and infrastructure, versus AgentsPodium's managed agent pods. ## FAQ ### Is there an AgentsPodium MCP server? You don't need an MCP server to give your agent AgentsPodium. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls the deploy, configure, lifecycle, and payment operations. The API itself can also set which MCP servers a deployed pod enables, configured through the tools operation. ### Can I limit what my agent is allowed to do with AgentsPodium? Yes. Write a rule that allows the pod read and configuration operations for one agent id but not agent deletion, API-key minting, or crypto and Stars payment orders, so the agent can manage its own pod while your account and billing stay under your control, and every call it makes is logged. ### What authentication does the AgentsPodium account API use? Per its OpenAPI spec, the API uses a bearer token in the Authorization header: either an ak_live_ API key created on the account page, or a session token obtained by exchanging a six-digit e-mail code through the verify operation. Through Jentic the key is stored encrypted by your own Jentic One instance and injected only at call time. ### Can I deploy and pause an agent pod with the AgentsPodium account API? Yes. The create-agent operation deploys a new pod from a persona and engine, and the pause operation stops billing the pod's resources while keeping it; resume brings it back and rebuild restores it from the last backup. You can also read the pod's liveness and usage to decide when to pause or rebuild. ### What are the rate limits for the AgentsPodium account API? The OpenAPI spec does not specify request-rate limits; resource limits are instead tied to each agent's plan, and the usage operation reports current RAM and disk against that plan while the term operation reports when the plan ends. Check hosting.defispace.com for current plan and quota details. ### How do I deploy an agent pod with the AgentsPodium account API through Jentic? Call the create-agent operation with a persona and engine, install the model key and toolsets, then check liveness to confirm the pod is up. In Jentic, search for 'deploy an AI agent pod', store your API key once, and the agent runs the sequence. To run it on your own infrastructure, install Jentic One from its GitHub repo.