DSL for reusable, evaluable agent workflows
Problem
Developers want to abstract repeatable agent tasks into definable workflows to avoid full agent loops and enable independent evaluation of steps like evidence screening.
Pain points
- Need for a DSL that captures workflow steps and tool calls
- Desire to evaluate evidence screening independently
- Interest in reusable workflow definitions across projects
Questions to verify
- Would you use a DSL to define agent workflows instead of building full agent loops?
- Do you need the ability to evaluate evidence screening independently?
- Are you interested in reusable workflow definitions across different projects?
Evidence
fact
You can combine tool calls, code, Jev-powered system one decisions for things like routing and screening evidence, and agents when a step needs more investigation.
fact
The idea is to make the work easier to understand and avoid paying for a full agent loop where one isn’t needed.
quote
I'll study it as I am toying with "what should a workflow definition language look like".
quote
Graph workflows are useless if there are edges without cost or effort attributes and termination states without reward.
quote
Interesting use of Jev-as-a-judge. How do you deal with having to reason before judging a step?
Disclaimer: this card is distilled by AI from community discussions as “pain point evidence”. It is a starting point for need analysis and validation, not a “validated market need”. AI has analytical limits; please open the original discussion above to manually verify before key decisions.