| name | cave-dovetail |
| description | Hand off specific named data from one agent to the next, typed (load files, validate expected outputs, map outputs→inputs). Use when step B needs particular named inputs produced by step A, not just the raw text. `dovetail()`. |
Dovetail — typed hand-off of named data between agents
The typed joint between two steps: a >> dovetail(D) >> b applies a DovetailModel between them — declares the expected outputs of the previous step, loads files into the context, and maps them to the next step's named inputs. A missing expected output surfaces as a BLOCKED/ERROR Link.
Native API (how you program)
from cave_teams import dovetail, DovetailModel
D = DovetailModel(expected_outputs=["spec"], file_inputs={"plan": "PLAN.md"})
piped = a >> dovetail(D) >> b
Via cave() (drive it from data)
{"op": "dovetail", "a": <spec>, "b": <spec>,
"model": {"expected_outputs": ["spec"], "file_inputs": {"plan": "PLAN.md"}}}
If the spec is malformed, cave() returns {"status": "construction_error", "hint": "read the cave-dovetail skill"} — the error names this skill.
See also
cave-sequential · cave-dag · cave-teams
Part of the cave-teams plugin — the language + full DSL is the cave-teams skill; drive any pattern from data with the cave skill.
