A2A
An open protocol that lets independent agents discover each other and collaborate across frameworks and vendors.
The plumbing — protocols, orchestration, and tool layers agents run on.
8 tools in this category
An open protocol that lets independent agents discover each other and collaborate across frameworks and vendors.
A tool layer that wires agents into hundreds of real apps and APIs with managed authentication.
One OpenAI-compatible interface to 100+ model providers. Swap or load-balance across OpenAI, Anthropic, Bedrock, Vertex, and local models without changing your code — and run it as a proxy that tracks spend, sets budgets, and manages keys.
Easy to setup, great to manage llm endpoints from different providers. Nice for first level tracing/observability (tracking costs and budget of deparments for example) or token cost optimisation via headroom as u set it up once and it works for the whole company.
The official directory of Model Context Protocol servers — a package index for agent tools.
A memory layer for agents. It distills conversations into durable, salient facts and recalls the right ones later, so an agent remembers across sessions instead of starting cold.
Tackles long-term agent memory by distilling salient facts — reports ~26% better recall with big latency and token-cost savings.
Turns a website's own content into a conversational, agent-readable interface.
A multiplayer agent harness for teams. One shared core gives every person their own isolated workspace — scoped memory, files, keychain, and a durable sandbox — in Slack and on the web, and it runs on any harness (Pi, OpenCode, Codex, Claude Code) so you're not locked to a vendor.
A durable execution engine that keeps long-running agent workflows reliable through retries, failures, and restarts.
Considering it to lean on managed services; note it isn't widely known yet (Dennis hadn't come across it).