test(e2e): add subagent-delegation task forcing a task sub-agent#3725
Merged
Conversation
The committed e2e suite has no task that reliably makes the model delegate, so it never exercises the sub-agent path end-to-end. This task seeds three files with arbitrary numbers and instructs the agent to read them via a single `task` sub-agent (not inline), then write their sum to result.txt. Because e2ebench drives the headless `reasonix run` path, this covers a fresh `task` delegation there — the exact path that regressed when persisted sub-agent transcripts made a parent session mandatory. continue_from/fork_from need a persisted session and are out of scope for the headless harness.
99be7c3 to
15d215c
Compare
SuMuxi66
pushed a commit
to SuMuxi66/DeepSeek-Reasonix
that referenced
this pull request
Jun 10, 2026
esengine#3725) The committed e2e suite has no task that reliably makes the model delegate, so it never exercises the sub-agent path end-to-end. This task seeds three files with arbitrary numbers and instructs the agent to read them via a single `task` sub-agent (not inline), then write their sum to result.txt. Because e2ebench drives the headless `reasonix run` path, this covers a fresh `task` delegation there — the exact path that regressed when persisted sub-agent transcripts made a parent session mandatory. continue_from/fork_from need a persisted session and are out of scope for the headless harness.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Adds an e2e suite task (
benchmarks/e2e/tasks/subagent-delegation) that reliably forces the agent to delegate via thetasktool, so the suite actually exercises the sub-agent path end-to-end against the real provider.workdir/data/{alpha,beta,gamma}.txtwith arbitraryname=numberlines (17, 28, 41).tasksub-agent (not inline) and writing their sum toresult.txt.verify.shgradesresult.txt == 86.Why
The existing suite (compaction / fix-add-bug / fizzbuzz / palindrome) never reliably triggers delegation, so it gives no signal on the sub-agent machinery. Because
e2ebenchdrives the headlessreasonix runpath, this task covers a freshtaskdelegation in headless mode — the exact path that regressed when persisted sub-agent transcripts made a parent session mandatory (fixed in #3586). The arbitrary numbers mean the answer can only come from actually reading the files; the run log shows thetaskinvocation as corroboration.Scope / notes
continue_from/fork_fromneed a persisted session, which headlessrunnever has, so they are intentionally out of scope here (they correctly error in that path). Those still need manual TUI/serve/desktop verification.main-v2), so this must land here to be picked up by/e2e. Until Continue subagent transcripts #3586's fix reachesmain-v2, this task will fail for binaries built without the ephemeral fallback — which is the intended regression signal (accuracy is reported, not a hard gate).Test plan
task.tomlparses with the harness'sBurntSushi/tomlloader (no unknown keys).86(incl. surrounding whitespace) → pass;85/99/missing → fail.go build ./cmd/e2ebenchgreen./e2eafter merge.