Turn-taking
The flagship turn-taking API is described as handling the rhythm of real conversation, including when to speak, wait, or interrupt, with norms, theory of mind, and memory composed underneath.
Humalike provides HTTP APIs for building humanlike AI agents that can time responses, adapt to group norms, remember people, and read how messages may land. It is aimed at developers building agents for chats, communities, games, support, care, and companion experiences.
Humalike is a behavioral infrastructure platform for humanlike AI agents. It provides HTTP APIs that help agents decide when to speak, how to fit into a group, what they should remember about people, and how a message may land before it is sent.
The product is positioned as model-, use-case-, and stack-agnostic, with support for both one-to-one interactions and group settings. Its documentation centers on a shared API pattern: authenticate with a bearer token, send JSON to the relevant endpoint on api.humalike.com, and read back a structured JSON response.
The flagship turn-taking API is described as handling the rhythm of real conversation, including when to speak, wait, or interrupt, with norms, theory of mind, and memory composed underneath.
Theory of mind is meant to help an agent infer what people think and feel as a conversation unfolds, so replies can be grounded in the interaction rather than just the last message.
The norms API reads a group’s tone and adapts the response to what is accepted in that specific conversation or community.
Persona and social memory are designed to make agents more grounded in real community data, with personality that can take sides and memory for who people are and what matters to them.
Social signals and social observability focus on cues and reception: pauses before sending, deleted reactions, edited messages, engagement, boredom, and annoyance.
The docs describe a consistent HTTP API pattern across the platform: bearer-token authentication, JSON requests and responses, one shared base URL, and a shared credit balance.
Build NPCs, teammates, or opponents that respond in rhythm with the conversation, remember the player, and feel less like a static prompt box.
Add behavioral layers to agents that own tasks, join channels, brainstorm, and respond in the tone of the team rather than as a generic assistant.
Support mental-health or care workflows where an agent needs to notice distress signals, maintain patient context, and know when to escalate to a human.
Create hardware companions such as pendants, speakers, or glasses that remember context, sense mood, and know when to speak or stay quiet.
Power community-facing bots that can joke, hype, react socially, and keep online spaces active without feeling detached from the group.
Humalike provides a suite of HTTP APIs for building agents that talk, time, and read people like a human. The docs describe one shared API pattern across the product: authenticate with a bearer token, send a JSON request to the relevant endpoint, and read a JSON response.
The docs show a standard quickstart flow: obtain a bearer token, set it as an environment variable, then send a POST request to the API endpoint with a JSON transcript body. The quickstart uses the Social Learning API as the example, and says the other APIs follow the same pattern.
The pricing page lists a free Developer plan, a Pro plan for production applications, and an Enterprise plan for larger deployments. It also says new accounts get $20 in free credits with no card required.
The product is built around behavioral APIs such as turn-taking, theory of mind, norms, persona, social memory, social signals, and social observability. The homepage says these APIs are model-, use-case-, and stack-agnostic and built for 1:1 and group settings.
The homepage and use case pages position it for AI gaming characters, coworkers, therapy or care agents, companions, ed-tech, humanoids, and creators or community bots.
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
AakarDev AI helps teams manage AI provider access, project-level setups, logs, and analytics from one dashboard. It supports BYOK workflows and lists providers including OpenAI, Google Gemini, Anthropic, Groq, Mistral AI, and Perplexity AI.
AgentMail is an email inbox API for AI agents that lets developers create, send, receive, and search messages through REST APIs and SDKs. It supports agent workflows such as threaded replies, verification, customer support, scheduling, and inbox-based approvals.
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
Lasso is an ecommerce product data platform for enriching catalog records, processing supplier files, generating product content, and monitoring competitors. It combines a web app with a REST API, SDK, and MCP server for teams and developers.
Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI. They help extend Codex with connected-service tasks, reusable instructions, and shared team workflows.