AI agents

Connect an AI agent

Give Claude, Hermes, OpenClaw, or another agent access to your AlertForge account. Once connected, it can plan, create, monitor, and export animated overlay packs on your behalf — spending your own account's credits, the same wallet the web app uses.

Endpoint

AlertForge's agent surface is one Streamable HTTP MCP endpoint: https://alertforge.ai/api/mcp. Unauthenticated calls reach three read-only info tools. Everything that touches your account requires an Authorization: Bearer <key> header, using a key minted at /dashboard/agents.

Set up your agent

Create a key at /dashboard/agents, then follow the steps below for your agent.

Give Claude a tool connection (MCP) to your AlertForge account.

1
Open Claude, then go to Settings and select Connectors.
2
Select "Add custom connector".
3
Paste this address into the URL field:
Connector URL
```
https://alertforge.ai/api/mcp
```
4
Open "Request headers" and add one: name it authorization, and for the value type Bearer <YOUR-KEY> (with a space after Bearer).
5
Select "Add". Claude can now build overlay packs when you ask it to.

Request headers is a new option Anthropic is still rolling out. If you don't see it yet, your account hasn't received it.

Tools

Three tools need no key. The other eight act on your account and require the bearer header above.

Public — no auth

`get_product_info`	What AlertForge is, its key features, and platform/OBS compatibility.
`get_pricing`	Current subscription tiers and one-time packs, sourced live from billing config.
`get_faq`	A handful of real, frequently asked billing and product questions with their answers.

Account — key required

`list_scene_kinds`	Which overlay scene kinds exist right now, and which ones your agent can create.
`list_particles`	The particle effects available to attach to a scene.
`get_balance`	Your current credit balance and a top-up link.
`list_overlay_projects`	Your existing overlay packs, with editor links.
`plan_overlay_pack`	Prices a pack spec and returns the exact credit cost before anything is created.
`create_overlay_pack`	Creates the pack from a priced plan; returns an editor URL immediately.
`get_pack_status`	Advances the pack and reports per-scene progress — call this in a poll loop.
`export_pack_zip`	Packages finished scenes into a ZIP and returns a signed download link.

How a pack gets built

plan_overlay_pack prices your spec and returns the exact credit cost before anything is created. create_overlay_pack spends that plan and returns a project and editor URL right away. get_pack_status advances the pack and reports progress per scene — an agent calls it in a loop until every scene is done. From there, either export_pack_zip for a downloadable ZIP, or hand the editor URL back to the user to finish in the browser.

Credits

Generation spends credits from your AlertForge account — the same balance the web builder draws from. plan_overlay_pack quotes the exact cost up front, and create_overlay_pack won't spend anything your balance can't cover. See /pricing for what credits cost and how each tier allots them.

Limits

Revoking a key at /dashboard/agents takes effect immediately — the next call fails. Each key is rate-limited (both overall and specifically on create_overlay_pack) and capped on how many new packs it can start per day. Both limits exist to catch a runaway agent loop, not to constrain normal use.

Export vs. the full pack download

export_pack_zip gives you the flat agent pack: each scene's WebM/PNG plus a manifest.json, zipped with a signed download link. The full themed folder download — the same canvas-composited branding and scene/stinger/alert layout the web app's own "Download pack" button produces — isn't part of the agent pack. Every tool result includes the editor URL; open it to get that from the web app.

/docs/apiAPI & agent access — the unauthenticated MCP tools and machine-readable discovery documents
/.well-known/agent-skills/index.jsonAgent Skills index — step-by-step skills for building a pack and reading pricing
/llms.txtllms.txt — whole-site fact sheet for AI agents