WIP: Integrate mini-SWE-agent as external baseline - #6
Open
bigboateng wants to merge 1 commit into
Open
Conversation
Add MiniSWEAgentAdapter that spawns the mini CLI as a subprocess, parses trajectory output, and verifies results with the same test/typecheck verifiers used by other agents. Runner changes: add "swe" agent type, weighted concurrency pools for heavy/light jobs, CLI filters (--tasks, --agents, --levels), dotenv support. Known remaining work: - Trajectory metrics parser does not yet extract steps/tokens from mini v2 format (messages key instead of history) - Debug shape logging still in adapter (remove after parser is fixed) - Default timeout bumped to 300s but may need tuning per task Made-with: Cursor
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
MiniSWEAgentAdapterthat invokes the mini-SWE-agent CLI as a subprocess, providing a recognized external baseline alongside the existing baseline and ZCA agents--tasks,--agents,--levels), anddotenvsupport for.env-based API key loadingMSWEA_CONFIGURED=1to skip mini's interactive setup wizard when spawned headlessKnown remaining work
messageskey, nothistory)Test plan
tsc --noEmitpasses cleanmini-swe-agent:parser_bug— PASS, exit code 0, verification confirmedconfigs/benchmark-with-swe-agent.jsonMade with Cursor