dongarandClaude Opus 5.5 659915d309 Fixes from the first live session of the new loop ...

Found live on 2026-10-07 (bring-up on nisse-testing-operator/nisse-bringup,
Opus as driver):

- The driver's prose nudge went into the system prompt, so the retry ended
  on the model's own reply. Opus 5.5 refuses that ("does not support
  assistant message prefill. The conversation must end with a user
  message"). The nudge is now a harness fact in the conversation. The
  delegation harness's prose path did the same and is fixed likewise.
- A helper's first request ended with its own delegation_started event,
  rendered as an assistant message: every delegation would have failed on
  Opus. The helper now sees it as the harness handing over its task.
- model_contract checks that every request ends on the user's side, so
  scripted tests catch this from now on (each fix is mutation-checked).
- `status` (and `details`) in an item's thread were taken as replies: they
  woke the driver and cancelled a bug's countdown. They now answer at once
  and change nothing.
- The driver had no way to answer a remark without changing the item, so it
  answered in prose. Every run woken by a message (or a review comment) now
  offers `reply`, which answers where asked and returns to the very wait
  that was interrupted.

Also: scripts/bringup-repo.sh makes the bring-up's test repository
(make_repo/delete_repo, split out of fresh_repo), and made repositories get
the agent's labels (bug, discuss, design).

Tests: test_loop_live_fixes.py (one per finding); the walk sends `status`
into threads (status_changes_nothing) and its scripted driver sometimes
answers remarks with reply.

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

2026-10-08 00:02:50 +00:00

..

model_contract.py

Fixes from the first live session of the new loop

2026-10-08 00:02:50 +00:00

test_abandon.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_agent_build.py

Wait out LLM outages with backoff (L5: outages delay, never fail)

2026-10-07 21:11:45 +00:00

test_approval_binding.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_approvals.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_bench.py

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

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

test_canonical.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_catalog.py

Rename the delegate and consultant roles to helper and expert

2026-10-01 23:04:37 +00:00

test_config.py

Roles, capability profiles and help on the new loop (M2.6)

2026-10-07 16:33:26 +00:00

test_context.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_control.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_delegation.py

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

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

test_dev_main.py

Make the config file the only home for settings

2026-09-23 23:25:58 +00:00

test_dump.py

Give each role its own view of the transcript (M1.2)

2026-10-02 00:39:38 +00:00

test_evals.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_execution.py

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

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

test_executor.py

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

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

test_expert.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_gates.py

Retire the old build spec, and point every reference at the architecture (M0.1)

2026-10-01 00:17:48 +00:00

test_git.py

Keep credentials out of reprs; count only live ssh-agents in tests

2026-10-02 22:41:01 +00:00

test_gitea.py

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

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

test_llm.py

Wait out LLM outages with backoff (L5: outages delay, never fail)

2026-10-07 21:11:45 +00:00

test_loop_conversations.py

Roles, capability profiles and help on the new loop (M2.6)

2026-10-07 16:33:26 +00:00

test_loop_help.py

Conversation on the new loop (M2.7)

2026-10-07 18:01:51 +00:00

test_loop_labels.py

Conversation on the new loop (M2.7)

2026-10-07 18:01:51 +00:00

test_loop_live_fixes.py

Fixes from the first live session of the new loop

2026-10-08 00:02:50 +00:00

test_loop_milestones.py

Roles, capability profiles and help on the new loop (M2.6)

2026-10-07 16:33:26 +00:00

test_loop_multi.py

Conversation on the new loop (M2.7)

2026-10-07 18:01:51 +00:00

test_loop_new_projects.py

New projects from conversation (M2.5, PR B)

2026-10-07 15:39:00 +00:00

test_loop_projects.py

Projects as configuration entries (M1.4, D8)

2026-10-06 01:12:57 +00:00

test_loop_properties.py

Fixes from the first live session of the new loop

2026-10-08 00:02:50 +00:00

test_loop_queue.py

Multi-milestone work and documents on the new loop (M2.4)

2026-10-05 20:57:37 +00:00

test_loop_talk.py

Conversation on the new loop (M2.7)

2026-10-07 18:01:51 +00:00

test_loop.py

Conversation on the new loop (M2.7)

2026-10-07 18:01:51 +00:00

test_matrix.py

Matrix threads in the transport (M2.1, PR A)

2026-10-02 23:40:24 +00:00

test_no_stranding.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_oplog.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_planner.py

Give each role its own view of the transcript (M1.2)

2026-10-02 00:39:38 +00:00

test_poller.py

Retire the old build spec, and point every reference at the architecture (M0.1)

2026-10-01 00:17:48 +00:00

test_pr_open_handling.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_pr_open.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_probe.py

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

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

test_publish.py

Publish only the work: leave build artifacts out and report them (#80)

2026-10-02 22:38:13 +00:00

test_quint_counterexamples.py

Flatten harness branch names so a model's branch can't block them

2026-10-02 22:29:20 +00:00

test_reconcile.py

Implement A9 state-change logging in the state store

2026-09-05 21:13:36 +00:00

test_render_anthropic.py

Add the Anthropic passthrough renderer and client (A10/A13) — 7a

2026-09-07 23:13:22 +00:00

test_render.py

Retire the old build spec, and point every reference at the architecture (M0.1)

2026-10-01 00:17:48 +00:00

test_review_loop.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_state.py

Implement A9 state-change logging in the state store

2026-09-05 21:13:36 +00:00

test_stop.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_tools.py

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

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

test_views.py

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

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

test_work_items.py

Put work items in the state store, and bind the log to them (M1.1)

2026-10-01 23:20:29 +00:00

test_worker.py

Add the operator kill switch, stop (A4)

2026-09-23 22:47:29 +00:00