podblocks
Run agents where it makes sense. Run code where you're comfortable. podblocks lets you use both in the same workflow, so they're as interchangeable as your work needs them to be. Technical or not, podblocks is built for you.
THE DIFFERENCE
Sometimes your agent just needs a prompt to carry out a task. Sometimes it needs a simple line of Python to transform your data or shape a response. We believe agents and code should live inside the same process, quickly and easily.
WHAT IS UNDER THE HOOD
HYBRID GRAPH
A pod is a DAG where each node is a typed block: a Python transform, a SQL statement, a browser script, a single agent, or a whole crew. They pass data through one typed artifact contract and run together. Most tools make you pick code or agents. podblocks runs both in the same graph.
CAUSAL POSETS
Each run produces a causal event poset following our goRapide engine. This means we know not just what happened, but what caused things to happen, which makes the agent smarter.
CONSTRAINT GATING
Declarative constraints over the causal stream (never, must_match) enforce five policies mid-run: halt, retry, route to a remediation block, require human approval, or log. Enforcement runs in-process on the worker at function-call latency.
CAUSAL OVERLAY
The UI overlays the runtime causal graph onto your design-time DAG. Crew nodes expand into their internal subgraph of agent, task, tool, and LLM events. Click any event and the inspector answers root-cause and descendant queries.
TRIGGERS & RESIDENT PODS
Pods run to completion or stay resident and react continuously. Triggers fire on a schedule, a signed webhook, a manual click, or a causal pattern over the live event stream. A violation in one pod can wake a resident remediation pod, and the combined record renders as one connected graph.
PRICING
It's free for 30 days, then the service is just $14.99 per month for single users. Team pricing is available for teams that need to build agents together.
HOW IT WORKS
Draw a DAG on the canvas. Each node is a typed block: Python, SQL, browser, agent, or a whole crew.
The control plane ships a sealed bundle to a worker, which compiles the pod to a CrewAI Flow and runs it.
Every block, agent, tool, and LLM event lands in a causal poset, with the edges that say what caused what.
Constraints watch the causal stream and act mid-run: halt, retry, route, require approval, or log.
BUILT IN GO
podblocks is built in Golang, so it's fast. It runs on the goRapide engine, so every agent event carries causal telemetry that is auditable and queryable, which means you can understand exactly why your agents pass or fail their tasks.
Watch a demo of podblocks in action, and see how a task gets accomplished quickly using a combination of agents and code in inline blocks.