fix: BUG: @hasna/codewith — native loops cannot represent a non-expiring

[andrei-hasna]

Objective

BUG: @hasna/codewith — native loops cannot represent a non-expiring schedule

Triggered by Chief Knowledge coordination task ac9afc48-6357-46cf-879a-c2ebd5891565 and fixed reviewer NO_GO. Repro: manage_loop create always persists a finite expiresAt (default seven days; explicit 4102444800 remains finite), while native readback exposes no non-expiring/null mode and no runtime-owned automatic renewal receipt. This makes the rendered permanent-ten-minute-coordination acceptance gate unreachable. Repair the owning Codewith scheduler with the smallest supported contract: either a true non-expiring recurring loop or a runtime-owned automatic create-first renewal that proves a replacement firing before expiry with no zero-carrier interval. Preserve existing finite-expiry behavior for callers that request it. Required delivery: focused regression tests, PR-first review, release/install, and live manage_loop create/list proof showing the permanent carrier cannot expire; then the same reviewer lineage on ac9afc48-6357-46cf-879a-c2ebd5891565 rechecks only this defect and direct regressions. Do not contact or probe agent-chief-planning.

Verification

• policy source: base 5b9d962 (immutable commit — agent-proof)
• containment: env — allowlist env, non-login shell, run-scoped HOME (registry auth seeded for install)
• install: pass
• doctor (ci): ok — 11 checks passed (1 advisory)

---

Run run_8c3826e2e9e0 · backend codewith · task d60366e0-4d66-45a3-ab97-2b7fcc878d3b
🏭 Generated by @hasnaxyz/factory

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[andrei-hasna]

[REVIEW] GO — #558 @ d0c2178 — lens: correctness+security+gates, reviewer unresolved-account001 (1 of 1)

Reviewed candidate

• Verified HEAD d0c2178 against the supplied freshly fetched origin/main 9301a83.
• Read git log origin/main..HEAD, the complete two-file diff, the surrounding manage_loop handler and tool schema, and the scheduler persistence/claim/expiry SQL paths.
• The contract is coherent on reachable paths: omitted expires_at keeps the seven-day default, an integer keeps an explicit finite expiry, and null persists as SQL NULL; existing expiry and claim queries preserve and run NULL-expiry schedules.

Commands and gates

• bun install — exit 0; setup only, not a repo test gate; 521 packages installed, 0 setup failures.
• This repository declares no test gate in package.json. No test command was invented.
• This repository declares no typecheck script.
• gh pr view 558 --repo hasna/codewith --json statusCheckRollup — exit 0; 28 checks read: 25 SUCCESS, 3 SKIPPED, 0 failed, 0 pending.
  • Bazel test on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • cla — SUCCESS
  • Check for spelling errors — SUCCESS
  • Blob size policy — SUCCESS
  • cargo-deny — SUCCESS
  • build-test — SUCCESS
  • Build Codewith Linux CLI — SUCCESS
  • Detect changed areas — SUCCESS
  • python-sdk — SUCCESS
  • Bazel test on ubuntu-24.04 for x86_64-unknown-linux-musl — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 1/4 — SUCCESS
  • Format / etc — SUCCESS
  • sdks — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 2/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 3/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 4/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm (native main) — SKIPPED
  • cargo shear — SUCCESS
  • Argument comment lint package — SKIPPED
  • Bazel clippy on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • Bazel clippy on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • Verify release build on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • Argument comment lint - Linux — SUCCESS
  • Verify release build on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • Argument comment lint - Windows — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • CI results (required) — SUCCESS
  • [code]smith — SKIPPED
• THE LANE'S CHOSEN COMMAND: git diff --check origin/main...HEAD — exit 0; 1 command passed, 0 failures. This was not treated as a repository gate.

Blocking P0/P1 findings

• None.

Non-blocking follow-ups

• package.json has no test script; the forge's named build/test checks are the authoritative validation surface for this PR.

[andrei-hasna]

[REVIEW] NO_GO — #558 @ d0c2178 — lens: correctness+security+gates, reviewer unresolved-account001 (1 of 1)

Blocking P1 finding

• The required checks did not exercise the tree that would land on current main. After git fetch origin main, origin/main is 9301a83. After git fetch origin refs/pull/558/merge, the tested merge ref first parent is still 5b9d962. Comparing the tested merge-ref tree with the actual current-base merge tree returned TESTED_MERGE_TREE_MATCH_RC=1.
• The actual current-base merge tree creates cleanly and preserves both changed loop-control files byte-for-byte, but it also contains three newer main-side files absent from the tested tree. The code finding set remains clean; the blocker is that the required build/test gate is stale for the landing candidate.

Focused remedy

• Update the PR branch with current origin/main, run the staged secret scan before committing, push the new head, and rerun the forge checks. The same reviewer will recheck only this named gate defect, current-head identity, and direct regressions. This NO_GO is live only while d0c2178 remains the PR head.

[andrei-hasna]

Focused remediation pushed at a885d65.

• Merged current origin/main 9301a83 into the PR branch without conflicts.
• The staged secret scan reported no leaks and exited 0 before commit; the pre-push scan also exited 0.
• The merge commit uses subject chore: refresh PR base for current main and carries exactly one Agent: unresolved-account001 trailer.
• Push to factory/d60366e0-4d66-45a3-ab97-2b7fcc87-82449d4d exited 0. The prior NO_GO is stale once the PR head readback shows a885d65.

The same reviewer is now waiting for the new exact-head forge checks and will recheck only the stale-gate defect and direct regressions.

[andrei-hasna]

[REVIEW] GO — #558 @ a885d65 — lens: correctness+security+gates, reviewer unresolved-account001 (1 of 1)

Focused re-review of the named stale-gate defect and direct regressions

• Current origin/main is 9301a83 and is an ancestor of this head: FINAL_MAIN_ANCESTOR_RC=0.
• The tested GitHub merge ref first parent is the same current-main SHA, and its tree matches HEAD exactly: FINAL_TESTED_MERGE_TREE_MATCH_RC=0.
• The remediation changed neither reviewed loop-control file: LOOP_FILES_UNCHANGED_RC=0. The PR diff remains two files, 184 insertions and 10 deletions.
• Staged gitleaks scan before commit — exit 0, no leaks found. Pre-push staged scan — exit 0, no leaks found. Push to the PR branch — exit 0.

Commands and gates

• bun install — exit 0 on the reviewed change; setup only, not a repo test gate; 521 packages installed, 0 setup failures.
• This repository declares no test gate in package.json. No test command was invented.
• This repository declares no typecheck script.
• gh pr view 558 --repo hasna/codewith --json statusCheckRollup — exit 0 on this exact head; 28 checks read: 25 SUCCESS, 3 SKIPPED, 0 failed, 0 pending.
  • Bazel test on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • cla — SUCCESS
  • Check for spelling errors — SUCCESS
  • Blob size policy — SUCCESS
  • cargo-deny — SUCCESS
  • build-test — SUCCESS
  • Build Codewith Linux CLI — SUCCESS
  • Detect changed areas — SUCCESS
  • python-sdk — SUCCESS
  • Bazel test on ubuntu-24.04 for x86_64-unknown-linux-musl — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 1/4 — SUCCESS
  • Format / etc — SUCCESS
  • sdks — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 2/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 3/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm shard 4/4 — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm (native main) — SKIPPED
  • cargo shear — SUCCESS
  • Argument comment lint package — SKIPPED
  • Bazel clippy on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • Bazel clippy on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • Verify release build on ubuntu-24.04 for x86_64-unknown-linux-gnu — SUCCESS
  • Argument comment lint - Linux — SUCCESS
  • Verify release build on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • Argument comment lint - Windows — SUCCESS
  • Bazel test on windows-latest for x86_64-pc-windows-gnullvm — SUCCESS
  • CI results (required) — SUCCESS
  • [code]smith — SKIPPED

Blocking P0/P1 findings

• None. The previously named P1 is fixed and its affected gate passes on the exact current-base merge candidate.

Non-blocking follow-ups

• package.json has no test script; the forge's named build/test checks are the authoritative validation surface for this PR.