Repository navigation
Description
Problem
The Bats suites tests/skills/agent-commit/workflow.bats and base-merge.bats still test unattended-mode and base-merge shell behavior that agentrepo commit --unattended now owns, so they duplicate the Go tests and must shrink to the cases that still need a real shell or the fallback.
Part of #322, within #588. Child 09 in #322's Children list. Deletes test cases only; no skill source or generated skill changes.
Acceptance criteria
- AC1: Given every case in
tests/skills/agent-commit/workflow.batsandtests/skills/agent-commit/base-merge.bats, when the pull request is read, then each case is marked deleted with the Go test that covers it, or kept as a wrapper, fallback or real-shell case with the reason, and no case is unaccounted for. - AC2: Given the merged change, then every case marked deleted is absent and every kept case passes.
- AC3: Given the Go tests named for the deleted cases, when CI runs, then they pass on macOS, Linux and Windows.
- AC4: Given the generated skills, then nothing changes and
mise run check-generated-skillspasses.
Scope and non-goals
Scope: tests/skills/agent-commit/workflow.bats and tests/skills/agent-commit/base-merge.bats and a case ledger in the pull request.
Non-goal: changing any wrapper, fallback or generated output. Non-goal: removing the frozen fallback before the N-2 window closes. Out of scope: Bats suites for other skills.
Approach
Meet AC1 to AC4. List the cases, match each to a Go test from children 01 to 06, delete the duplicates and keep the wrapper and fallback cases. The mechanism is the implementer's choice.
Dependencies
Size
S. One pull request, deletions only. Estimated code SLOC counted with scc rules on the case files: 546 lines of code at most, under the 800 limit; none of it is in a generated path.
Verify plan
- AC1: the ledger in the pull request lists every case in
tests/skills/agent-commit/workflow.batsandtests/skills/agent-commit/base-merge.bats. - AC2: the Bats run for the remaining cases passes and the diff shows only ledger-marked deletions.
- AC3: the CI matrix jobs for the named Go tests.
- AC4:
mise run check-generated-skillspasses with no diff. - Definition of done: D1 each AC is verified at the merged head, with the commands and results cited in the pull request's Verification. D2 an independent review approves the exact merged head, and every finding has a recorded disposition. D3 required checks pass with no gate loosened. D4 the leak scans the repository specifies are clean, and every post made for this work passes the per-post leak review before it is posted. D5 every commit is signed. D6 the documentation matches the change or the Verification says why there is no impact. D7 every commit has an
Assisted-bytrailer and the pull request has theprovenance:agent-assistedlabel. D8 the pull request stays within the size limits. D9 follow-ups are filed as issues and listed under Related. D10 the merge closes the issue through a closing reference, the board shows Done, and the finalStatus:comment names the merge commit.
Risks
Deleting a case whose Go equivalent is weaker loses coverage; the ledger names the covering test and the review checks it. Undo: revert the change, which restores the cases.
Operator authority
None.
Metadata
Metadata
Assignees
Labels
Type
Fields
Priority
Work type
Projects
- StatusShow more project fieldsBacklog
Activity
tbhb-agent commentedon Oct 27, 2026
Refinement: READY — Retire the agent-commit Bats suites where Go covers them
Model: gpt-6-astra
Child 09 Retire the agent-commit Bats suites where Go covers them
Verdict: READY
Checked: The unchanged suites total 603 physical lines, bounding this deletion-only scope below 800. The ledger includes redesign and merge-parent verification behavior represented at
tests/skills/agent-commit/workflow.bats:328andtests/skills/agent-commit/base-merge.bats:262.AC1–AC4 have matching checks. Dependencies supply replacement behavior and wrapper parity; retained shell cases remain permitted. Scope, authority, risks and undo are consistent.
Gaps: None.
Split: None. S remains credible.
Definition of done
The definition of done for this issue (Chore). Tick each item and give its evidence.
check-leaks.shwith the repository's private-strings list, and the leak review of each post.docs/, CHANGELOG or help text), or the Verification section says "No doc impact" and why.provenance:agent-assistedlabel.Status:comment names the merge commit.mise run checkis green, and so are the coverage floorsmise run cover-go-checkandmise run cover-py-check, whichcheckdoesn't run.packages/agents/source/commits whatmise run generate-skills, thenmise run skill-tokens, thenapm installrewrite, in that order, andcheck-generated-skills(part oflint) passes.docs/agent-skills-migration.mdor the adoption docs, or the PR's Verification says "No consumer impact" and why.Round: 3 of 3