dongarandClaude Opus 5.5 37965aefca Fix the run-breaking tool-result order; e2e test runs the slice (M2.1, PR D) ...

The first live run of the new loop's e2e test found that a fact recorded
while start_milestone was in flight landed between the tool call and its
result; both API formats reject that, so every later request in the run
failed. Views now keep each result right after its call. Scripted test
models check the APIs' ordering rules, so this can't pass unit tests
again.

The e2e tests no longer depend on the sandbox's history: each run asks for
a uniquely named function and works against a base branch of its own,
deleted afterwards. The review step asks for a change that can't already
be there. The scripted operator gives up after five fruitless answers, and
each run keeps its state store for inspection. The driver's prompt says
review fixes stay in the current milestone.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPQ14L3LikyefCMFBqEbKE

2026-10-03 15:11:06 +00:00

..

benchmark.py

Run every command on the build VM over SSH (M1.3)

2026-10-02 21:39:46 +00:00

bringup.sh

Run every command on the build VM over SSH (M1.3)

2026-10-02 21:39:46 +00:00

check.sh

Add the end-to-end test, run after each milestone (M1.6)

2026-10-02 22:59:13 +00:00

e2e.sh

Fix the run-breaking tool-result order; e2e test runs the slice (M2.1, PR D)

2026-10-03 15:11:06 +00:00

evaluate_model.sh

Add an executor benchmark that keeps its results

2026-09-25 00:35:24 +00:00

probe_context.py

Enforce the tool-call contract, asking servers only where it works

2026-09-24 22:54:49 +00:00