fix(flywheel): phase-1 repairs for broken flywheel loop - #41

Merged

robotlearning123 merged 16 commits into

agent-next:mainfrom

robotlearning123:pr/flywheel-phase1

Mar 5, 2026

Commits on Mar 5, 2026

feat(pipeline): autonomous pipeline v0.1.1-v0.1.3

robotlearning123
and
claude
committed
76e3d70
fix(pipeline): memory leak, stale ctx, redundant map lookups

robotlearning123
and
claude
committed
a27bd6b
feat(pipeline): error context retry, wave validation, budget-based loop (v0.1.4)

robotlearning123
and
claude
committed
1d0fa91
task(f816b860e3c946ac): feat(classifier): add classifyTask pure function
robotlearning123
committed
d3acd53
task(b24b8f37e8d64ede): feat(types): add optional model field to Task
robotlearning123
committed
c7235e5
task(654d29040498495e): feat(agent-runner): pickFallbackAgent + per-task model override
robotlearning123
committed
8276adb
feat(scheduler): task classifier + model fallback on retry (v0.1.5)

robotlearning123
and
claude
committed
ca855f3
task(6f72353ece8b4c67): feat(worktree-pool): add getActiveWorkers() and rebaseOnMain()
robotlearning123
committed
66bb12b
task(6986f62d76ef4228): feat(store): persist _originalPrompt; serialize dependsOn array as JSON
robotlearning123
committed
721084d
task(71e20b9544914d85): test(scheduler): prompt accumulation fix, model escalation, array
robotlearning123
committed
34fcff9
feat(scheduler): prompt accumulation fix, model escalation, staged rebase, multi-dep DAG (v0.1.6)

robotlearning123
and
claude
committed
b4d29d2
feat(v0.1.7): empty commit detection, GPT-5.4 routing, pricing table, session resume, codex config

robotlearning123
and
claude
committed
265e914
docs: add strategy, research, plans, and pipeline artifacts

robotlearning123
and
claude
committed
35a59b1
docs: fix stale versions, model names, honest flywheel status

robotlearning123
and
claude
committed
f0f0e23
fix(flywheel): phase-1 repairs for broken flywheel loop

robotlearning123
and
claude
committed
6d8f462
fix(test): add git config for CI in pipeline full-flow test

robotlearning123
and
claude
committed
b613481

Navigation Menu

Uh oh!

fix(flywheel): phase-1 repairs for broken flywheel loop - #41

fix(flywheel): phase-1 repairs for broken flywheel loop#41
robotlearning123 merged 16 commits into
agent-next:mainfrom
robotlearning123:pr/flywheel-phase1

Commits on Mar 5, 2026

feat(pipeline): autonomous pipeline v0.1.1-v0.1.3

fix(pipeline): memory leak, stale ctx, redundant map lookups

feat(pipeline): error context retry, wave validation, budget-based loop (v0.1.4)

task(f816b860e3c946ac): feat(classifier): add classifyTask pure function

task(b24b8f37e8d64ede): feat(types): add optional model field to Task

task(654d29040498495e): feat(agent-runner): pickFallbackAgent + per-task model override

feat(scheduler): task classifier + model fallback on retry (v0.1.5)

task(6f72353ece8b4c67): feat(worktree-pool): add getActiveWorkers() and rebaseOnMain()

task(6986f62d76ef4228): feat(store): persist _originalPrompt; serialize dependsOn array as JSON

task(71e20b9544914d85): test(scheduler): prompt accumulation fix, model escalation, array

feat(scheduler): prompt accumulation fix, model escalation, staged rebase, multi-dep DAG (v0.1.6)

feat(v0.1.7): empty commit detection, GPT-5.4 routing, pricing table, session resume, codex config

docs: add strategy, research, plans, and pipeline artifacts

docs: fix stale versions, model names, honest flywheel status

fix(flywheel): phase-1 repairs for broken flywheel loop

fix(test): add git config for CI in pipeline full-flow test