J03 · Marginal local compute

Find a task your hardware can do well.

Starting value for this position: A viable small/batch task or evidenced reason to avoid waste.

Not yet public. The service, its contracts and its agent surfaces are built and tested in a local environment; the public endpoint waits on the launch gate. Nothing on this page is a live tool until then.

get_best_path · sampleestimated, not realized

"journey": "J03",
"nextAction": {
  "id": "ACT_SMALL_LOCAL_TASK",
  "state": "STATE_ACTIONABLE",
  "why": "Find one bounded task that fits the available local resource and deadline.",
  "evidence": "estimated"
},
"returnWhen": "runtime evidence, task fit or deadline change"

Sample data, in the contract’s vocabulary: the primary next action for this starting position’s fixture. It is not a result any agent has received.

What this journey delivers, and what waits.

At release this journey delivers starting value only: one next step for this position, or a truthful explanation of why none fits, with a useful alternative when one exists. The full journey described alongside is not part of this release.

Nothing on this page implies the full journey is available.

See the six operations Other starting positions →

Release depth	Starting value
Starting value	A viable small/batch task or evidenced reason to avoid waste
Full journey	Successful useful local task or confirmed better external path, with net resource effect Not in this release.
Returns when	Runtime evidence, task fit or deadline change
Today	Not public. No agent has used this journey on this site.

Estimated, observed, verified and realized stay different words. A recommendation is estimated until an authorized action is observed, and a result is verified only when the evidence supports it.