lab: wfbench — Workflow Editor Agent Benchmark harness + core pack step (re-land on main) - #1651
Merged
Conversation
…3 twin) (#1650) * lab: wfbench — Workflow Editor Agent Benchmark harness (stakwork 58313's twin) wfbench-run: graph roster (EvalSet / EvalRequirement / EvalTrigger with 58313's ids, HAS_BASELINE_TRIGGER on the first trigger) in parallel with the meta/* author agent -> resolve the shipped version -> input-key gate -> rerun via meta/run-workflow -> per-criterion judge subflow -> eval/aggregate-scores -> record EvalTriggerOutput + CriterionResult per 58312 -> one Hive callback body (POSTed and returned). Ten pure steps, two workflows, seeded unstamped; wfbench/* is never an agent tool. Offline smoke (src/lab/wfbench/smoke.ts): seeds, discovers, static- validates both workflows through the authoring capability, drives every pure step, and checks every graph payload's attributes and edges against JARVIS_ONTOLOGY. * vein: core `pack` step; lab: drop the three per-experiment pack-result copies A workflow's output is its last step's output, so returning fields from several steps needs a passthrough step — harvey, gaia and wfbench each carried an identical one, and the wfbench author reached for gaia/pack-result because it was the first it found. `pack` is that primitive in core (config = output), discoverable everywhere. Every harvey/gaia/wfbench workflow, seed, smoke and doc is re-pointed; the copies are deleted. (cherry picked from commit dbc98d5)
This was referenced Sep 4, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Re-lands #1650 on main. #1650 was merged into its stacked base branch (
meta-validate-and-eval-promote) after #1648 had already merged that branch into main, so its squash commit never reached main. This is that commit cherry-picked onto current main; nothing new.Contents (see #1650 for the full description):
mcp/src/lab/wfbench/(10 pure steps,wfbench-run+wfbench-judge-criterion, offline smoke), the vein corepackstep replacing the three per-experimentpack-resultcopies, lab wiring + docs.Verified on #1650: vein
npm test665/665; mcp tsc clean; wfbench / harvey / gaia smokes green; one live run 4/4 with the graph chain verified in Neo4j.