Research-led AI infrastructure
Sparkco presents itself as an open-source research project on the post-AGI stack, focused on runtime containers, harnessing, and messaging for AI agents.
Sparkco: infraestrutura open source para agentes de IA, ferramentas focadas no terminal, fluxos para mercados de previsão e planos de agentes de voz por uso com crédito inicial, SMS, transcrições e integrações.
Sparkco is an open-source research project and product site focused on the infrastructure layer for AI agents. The homepage describes three core areas: runtime containers that agents live in, the harnessing or glue code inside them, and the messaging between them.
The site also frames Sparkco as a public log of active research and tooling, with live AI and prediction-market signals, curated recommendations for agent builders, and open code. Its blog content and open-source listings show a strong emphasis on terminal-based workflows, especially around prediction-market monitoring and trading.
The pricing page shows a separate commercial offering built around voice handling, with minute-based plans, SMS updates, dashboards, transcripts, and API/webhook access on higher tiers. Together, the site positions Sparkco as both a research publication and an operational product for teams that need agent infrastructure or call-handling automation.
Sparkco presents itself as an open-source research project on the post-AGI stack, focused on runtime containers, harnessing, and messaging for AI agents.
The site says the team ships tools as CLIs first, with a stated preference for terminal workflows over MCP exposure.
The public site acts as a live log of AI and prediction-market signals, along with setups and tools recommended for agent builders.
The open-source section lists public projects such as harness/agent runtimes, memory testing, autonomous agent arenas, and trading tools.
The pricing page offers voice-agent plans with minute-based billing, SMS updates, live analytics, transcripts, API and webhooks, and custom integrations on higher tiers.
The pricing page includes add-ons such as analytics and transcript search, warm-transfer minutes, and a dedicated success engineer for larger deployments.
Teams building AI agents can use Sparkco’s public framing and tooling examples to think about runtime containers, harnessing loops, and agent-to-agent messaging.
Operators running voice workflows can compare usage-based plans, SMS updates, dashboards, transcripts, and support tiers before starting a pilot.
Developers who prefer terminal workflows can use the CLI-first material and open-source projects as references for building deterministic harnesses and shell-based workflows.
Prediction-market traders and builders can follow the blog’s terminal dashboard and mispricing examples to monitor markets and structure command-line workflows.
You can create an account without a credit card and receive $20 of voice credit to start. The pricing page says that is enough for roughly 40 minutes of calls at the Starter rate or a mix of calls and SMS.
Only live agent-to-caller time is metered. Hold music, warm-transfer time, and outbound texts are not counted, and billing is by the second.
The pricing page describes Starter, Growth, and Scale tiers, plus add-ons such as analytics and transcript search, warm-transfer minutes, and a dedicated success engineer. The home page also positions Sparkco as an open-source research project with CLI-first tools and public code.
The site lists HIPAA, TLS 1.2+ encryption, and SOC 2 in progress. It also offers a HIPAA policy white paper for download.
The homepage says Sparkco ships tools as CLIs first rather than MCP, and the blog examples show terminal-style workflows for prediction market monitoring and trading.
Os dados de tráfego são apenas para referência.
www.agentconnect.md
AgentConnect permite que as equipes integrem agentes ACP como Claude Code, Codex e Gemini CLI ao Slack, Telegram, Discord e GitHub, para que eles possam assumir tarefas nos mesmos canais em que as equipes já colaboram. É open source e pode ser hospedado por conta própria, com funções, ferramentas, memória e gatilhos configuráveis para fluxos de trabalho com agentes.
sim.ai
Sim é um workspace de IA de código aberto para criar, implantar e gerenciar agentes de IA, com fluxos visuais, Chat, acesso à API, integrações e ferramentas de colaboração.
dagworks.io
Ferramentas open source e hospedadas para fluxos de trabalho de IA confiáveis com Apache Hamilton e Apache Burr
chirpz.ai
A Chirpz AI desenvolve o PandaProbe, uma plataforma open source para rastrear, avaliar e monitorar agentes de IA em produção, aprimorando a observabilidade e a depuração dos fluxos de trabalho.
www.phinq.co
Phinq é uma camada open source de governança em tempo de execução para agentes de IA. Ela intercepta chamadas de ferramentas, classifica o risco e pausa ações de maior risco para obter aprovação humana antes da execução.
latitude.so
Latitude é uma plataforma de observabilidade e monitoramento open source para agentes de IA, que ajuda as equipes a inspecionar traces de produção, detectar problemas e verificar correções, com tracing baseado em OpenTelemetry, alertas, busca de sessões e fluxos de avaliação.