Workflows

Deterministic ticket steps, not LLM hope

Five dedicated ticket nodes make tracker operations deterministic steps in the DAG, no prompting an agent and hoping it calls the right tool. Transition a Jira ticket when a PR merges, comment the release note, and notify, in one canvas that also understands PRs, CI, and agents.

• ticket.create, ticket.update, ticket.comment, ticket.read, ticket.list
• Works across Jira, Linear, and ClickUp through one normalized interface
• Template variables wire trigger and step data into every field

5ticket node types

1. Visual builderReact Flow canvas with trigger, agent, git, CI, ticket, and approval nodes aligned to the 22px grid.
2. Reusable blocksSave a group of steps as a workflow block and reuse it across workflows.
3. Dry-run testingRun the whole graph against sample trigger data with side effects stubbed out.
4. AI generationGenerate a complete workflow (nodes, edges, and config) from a plain-English prompt.
5. Execution engineTopological scheduling, parallel branches, step retries, and live SSE streams.
6. Template engineReference trigger fields and previous step outputs with {{trigger.*}} and {{outputs.*}}.