fix(templates): stop the bulk archive when the user picks Cancel

[clay-good]

Status: LGTM — ready for review.

What was wrong

The generated openspec-bulk-archive-change skill (and the matching opsx:bulk-archive command) offered you a Cancel option at the confirmation prompt:

7. **Confirm batch operation**
   - Options might include:
     - "Archive all N changes"
     - "Archive only N ready changes (skip incomplete)"
     - "Cancel"

8. **Execute archive for each confirmed change**
   Process changes in the determined order...

Step 7 collected your answer, but nothing ever told the agent what to do with it. Step 8 followed immediately and archived every selected change. An agent reading the skill literally would mv your changes into archive/ after you explicitly cancelled — and the only trace of the option was a vague "Skipped: user chose not to archive (if applicable)" bullet three steps later, after the move already happened.

The single-change archive skill had the same hole; it was fixed in #1357 (46a4d78) and reworded into its current routing block by #1394. The bulk variant was left behind. Neither has shipped yet — both are post-v1.6.0.

How it was fixed

Both template bodies (skill and opsx command) now route the answer:

Route on the answer by intent, not by exact label — you wrote these labels,
so match what the user picked rather than the wording above:
- "Cancel" — stop, do not archive. Report that nothing was archived and skip the remaining steps.
- The archive-everything option — proceed with every selected change
- The ready-only option — proceed with only the changes the step 6 table marks `Ready` or `Ready*`, and record the rest as Skipped in step 8c. If a `Ready*` change's conflict partner is skipped, re-derive that conflict's resolution using only the changes being archived.
- Anything else — ask again rather than archiving

Four deliberate details, each from an adversarial review finding:

• Routing is by intent, not by literal label. Step 7 says options "might include" those labels, and they contain an N placeholder the agent fills in ("Archive all 3 changes"). Matching the literal strings would have sent every legitimate answer into "Anything else — ask again", trading a destructive bug for an infinite re-prompt.
• "Ready" is bound to where it's decided — the step 6 status table, including the conflict-flagged Ready* row. Otherwise the agent can't tell which subset to archive.
• A skipped conflict partner invalidates its recorded resolution. Ready-only can archive one side of a conflict while its partner stays active; without re-deriving, step 8a's "apply in resolved order" would sync the skipped change's deltas too.
• The Cancel wording matches the single-change skill verbatim ("Cancel" — stop, do not archive), so one assertion guards all four templates — and "skip the remaining steps" also rules out printing a "Bulk Archive Complete" summary after a cancel.

A guardrail line was added where agents skim: Never archive after the user cancels the confirmation — a cancelled batch archives nothing.

Proof it works

Two regression tests in test/core/templates/skill-templates-parity.test.ts:

• honors Cancel at every archive confirmation (#1381) — covers all four templates (bulk skill, bulk command, single skill, single command). The single-change routing has been correct since feat(skills): publish workflow skills to skills.sh #1357 but was never pinned by a test; this closes the whole issue rather than half of it.
• routes the bulk archive confirmation by intent, not by literal label (#1381) — pins the intent-based wording, the ready-only binding, and the guardrail.

Both verified failing-then-passing. With the template change reverted:

× honors Cancel at every archive confirmation (#1381)
  → bulk skill: expected '---\nname: openspec-bulk-archive-chan…' to contain '"Cancel" — stop, do not archive'
× routes the bulk archive confirmation by intent, not by literal label (#1381)
  → bulk skill: expected '---\nname: openspec-bulk-archive-chan…' to contain 'Route on the answer by intent, not by…'

Full suite after npm run build: 2029 passed / 2046, 104 of 105 files green. The sole failing file is test/core/completions/installers/zsh-installer.test.ts (17 failures) — the known environment-only oh-my-zsh issue noted in CLAUDE.md and tracked in #1321, untouched by this branch and green in CI.

Notes / nits

• Instruction text only. No CLI, parser, schema, or data-model changes; no workflow design or architecture change.
• Golden hashes regenerated for the three affected entries (getBulkArchiveChangeSkillTemplate, getOpsxBulkArchiveCommandTemplate, openspec-bulk-archive-change), plus the committed skills/openspec-bulk-archive-change/SKILL.md via npm run generate:skills. No other hash drifted.
• Sequencing: fix(templates): stop instructing a second date prefix on dated archive names #1388 touches the same file ~10 lines below this hunk and moves the same three hashes. The text auto-merges, but whichever lands second must rebuild and regenerate the hashes. feat(instructions): inject config rules into apply and archive #1062 has the same hash overlap.
• Heads-up: feat: support multi-level change domains and sibling archives #1367 is ~15 commits stale and its diff still carries Proceed to archive regardless of choice. in archive-change.ts. Because it edits those same lines, git will raise a conflict rather than silently revert — but resolving that conflict in feat: support multi-level change domains and sibling archives #1367's favor would reopen Generated openspec-archive-change skill archives even when the user selects Cancel #1381. The new four-template test now fails loudly if that happens.
• Not in scope: the bulk template's step 8a still describes the spec sync loosely ("Use the openspec-sync-specs approach") without the inline/wait guardrail fix(templates): don't archive a change before its spec sync finishes #1394 added for single-change archive. Worth a follow-up.

Closes #1381

🤖 Generated with Claude Code

Summary by CodeRabbit

• Bug Fixes
  • Bulk archive now properly honors “Cancel”—no changes are archived.
  • Confirmation routing is now based on selected intent (not exact label text): Archive all archives everything; Archive only ready archives only Ready/Ready* items.
  • Non-matching items are recorded as Skipped; if a Ready* item’s conflict partner is skipped, conflict resolution is re-derived from the archived set.
  • If the confirmation response is unrecognized, the user is re-prompted instead of proceeding, consistently across bulk and single flows.

[coderabbitai]

📝 Walkthrough

Walkthrough

Bulk archive instructions now stop on “Cancel”, proceed for the two archive options, and re-prompt for other responses. Generated template hashes and parity tests are updated accordingly.

Changes

Bulk archive confirmation

Layer / File(s)	Summary
Confirmation response routing .changeset/bulk-archive-honors-cancel.md, skills/openspec-bulk-archive-change/SKILL.md, src/core/templates/workflows/bulk-archive-change.ts	Bulk archive guidance explicitly stops on “Cancel”, archives all or only ready changes for the corresponding options, and re-prompts for unrecognized responses.
Generated template parity validation test/core/templates/skill-templates-parity.test.ts	Expected template hashes are updated, and parity tests verify cancellation, re-prompting, and intent-based routing text in generated variants.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• Fission-AI/OpenSpec#527: Updates the same bulk-archive template factories and confirmation behavior.
• Fission-AI/OpenSpec#1089: Updates the same bulk-archive workflow templates and guardrails.

Suggested reviewers: tabishb

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Linked Issues check	⚠️ Warning	The linked issue targets openspec-archive-change and opsx/archive, but this PR only updates bulk-archive templates and tests.	Update the archive-change skill and archive command templates so Cancel aborts before any archive step, or narrow the linked issue scope.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Out of Scope Changes check	✅ Passed	The changes stay focused on bulk-archive confirmation flow, template generation, hashes, and regression tests.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: bulk archive now stops when Cancel is selected.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/bulk-archive-honor-cancel

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[alfred-openspec]

The routing closes the destructive fall-through in both bulk templates: Cancel exits before sync or move, unknown responses re-prompt, and ready-only selection re-derives conflicts without skipped changes. All nine focused parity tests pass locally at 078a2f5 and the exact-head CI/CodeQL matrix is green, so this is approved.