Warden Studio

Automate publishing a Community Agent in Warden Studio through a safe, repeatable workflow that other agents can follow.

Safety & constraints (non-negotiable)

• Never request or store seed phrases / private keys.
• Never ask the user to paste secrets into chat. If an API key must be entered, instruct the user to paste it directly into the Studio UI field.
• Treat publishing/onchain registration as high-risk: confirm network, fees, and what is being signed before any wallet confirmation.
• Prefer read-only validation (checking forms, status, preview) unless the user explicitly authorizes execution (e.g., "yes, publish" / "yes, execute").
• Do not reveal any private info (local files, credentials, IPs, internal logs).
• Public comms: do not claim any affiliation or relationship unless it is publicly disclosed and the user explicitly asks you to state it.

What this skill does

Typical outcomes:

• Log into https://studio.wardenprotocol.org
• Create a new Agent submission/listing
• Provide:
  • API URL (service endpoint)
  • API key / auth method (if required)
  • Name, description, skills, avatar
  • Billing model (free vs paid per inference, in USDC)
• Pay registration fee + gas (if prompted by the UI)
• Verify the agent shows up in Studio and becomes discoverable in Warden's Agent Hub (Community tab), when applicable.

Workflow (UI automation)

0) Preconditions

1. A Chromium browser is available (Chrome/Brave/Edge/Chromium). (Firefox not supported.)
2. User can log in to Warden Studio (email/SSO/2FA completed).
3. The agent is already deployed somewhere and reachable via HTTPS (no UI required):
   • stable API base URL
   • (optional) API key or token if the endpoint is protected
4. Funding is ready for registration (if required by the flow):
   • USDC on Base for the registration fee (confirm the fee in the UI)
   • ETH on Base for gas

If any of the above is missing, stop and ask the user to do that step.

1) Open + stabilize Studio

• Open: https://studio.wardenprotocol.org
• Wait for the landing/dashboard to load.
• Take a snapshot and identify:
  • logged-in user / account handle
  • any "Agents" list/table or "Submit / Create agent" entry point
  • network/payment cues (e.g., Base, USDC, wallet connection state)

If Studio is gated by login, stop and ask the user to complete login in the UI.

2) Read-only checks (default)

Use these first to prevent failed submissions:

• Confirm the agent endpoint is reachable:
  • the URL is HTTPS
  • no obvious typos
  • (if a "Test connection" exists) run it
• Validate required metadata is prepared:
  • agent name (short)
  • description (clear, non-misleading)
  • skills list (concise + accurate)
  • avatar image ready (square recommended)
• Check billing/monetization options:
  • free vs per-inference (USDC)
  • expected fees shown by the UI

3) Draft the submission (no publishing yet)