fix(commands): task bookkeeping in implement, taskstoissues and converge

[ntdatt812]

Consolidates #4313, #4314 and #4330 into one pull request, as asked on #4460 (author-over-cap). Those three will be closed in favour of this one. Each also had review feedback outstanding; that is addressed here, and noted per section below.

The three share one subject: how the command templates read task bookkeeping in tasks.md and in the issue tracker. They stay separable. Each fix is its own run of commits, and each run was cherry-picked alone onto main and passes its own tests there, so any one can be dropped without touching the others.

Closes #4269, closes #4271, closes #4272.

1. /speckit-implement counts checkbox markers inside code fences (#4272), from #4313

implement.md defined its total, checked and unchecked counts on every - [ ] / - [x] line, so a checklist that documents the checkbox format in a fenced example reported items nobody can tick. A non-zero unchecked count stops implementation. /speckit-clarify already scoped its scan to markers outside code fences; implement now does the same.

tests/unit/test_checklist_scan_contract.py asserts that every line in a command template that defines a checkbox-marker scan also excludes code fences.

Review feedback: Copilot pointed out that the test only recognised a definition whose first marker is unchecked, so the "Checked items: Lines matching - [X]" definition was never parametrised; removing its exclusion left the suite green. Measured before changing it: with the old pattern that removal passes 4/4. The pattern now accepts a checked or unchecked first marker, four definitions are guarded instead of three, and the same removal fails.

2. /speckit-taskstoissues dedup matches bare task IDs across features (#4271), from #4314

Task IDs restart at T001 in every feature's tasks.md, and dedup matched existing issues on the ID alone. So the first feature to reach the tracker permanently suppressed T001 for every later feature, and those tasks were silently never created. Issue titles now carry the feature ([002-billing] T001: ..., from the FEATURE_DIR basename), and a task is skipped only when an issue matches both the feature and the ID.

Review feedback, and backward compatibility: Copilot found that the upgrade rule brought the bug straight back. It treated a bare legacy T001: ... title as this feature's whenever no scoped title existed for that ID, and that is the state of every feature on its first run after upgrading: a bare T001 filed for 001-auth still suppressed 002-billing's T001.

A bare title names no feature, and neither does the absence of a scoped one, so nothing in the tracker can settle which feature it belongs to. Deciding either way breaks someone: skipping drops a task silently, and creating duplicates one an existing user already tracks. That second case is the compatibility concern raised on #4314. So the command now lists every ID that matched only a bare title, each with the issue and this feature's description for the task, and asks before creating anything. Confirmed issues are skipped. It then offers to retitle them to the scoped form, and does so only for the ones the user agrees to, so the question does not come back on later runs. Existing users therefore get no duplicates and no silent gaps; the cost is one question per legacy issue, once.

The U+0008 bytes Copilot flagged on the earlier commit were already fixed in that PR (\b written as two characters); tests/integrations/test_integration_goose.py, which caught them, passes.

3. /speckit-converge is not idempotent (#4269), from #4330

#4330 dropped findings that an unchecked task already covered. Review there identified the root cause: converge never enforced its prerequisite. Run while tasks.md still had open tasks, it assessed the code anyway, found that unbuilt work as new gaps and appended it again under fresh IDs. And a run whose findings were all already tracked ended in converged, whose report says the implementation satisfies the spec while tracked work is still open.

Reworked around the lifecycle instead of the dedup rule:

• Step 1 enforces the prerequisite. If tasks.md has any unchecked task (outside code fences, the rule implement counts by), converge stops before assessing anything. It lists the open task IDs, points to /speckit-implement, and leaves tasks.md byte-for-byte unchanged. This is neither converged nor tasks_appended.
• Every task is assessed once all are checked. Tasks join the intent inventory, so one marked done but not built is a finding traced to its ID.
• Anything appended makes the next run stop until it is implemented, which is what makes a re-run safe. The dedup paragraph is removed, and converged is only reachable once the work is actually finished.
• The handoff and docs/reference/agentic-sdd.md describe the gate.

tests/unit/test_converge_prerequisite.py pins the lifecycle: the gate sits in Step 1 and stops before Step 2, names implement, lists the open IDs, leaves tasks.md untouched, counts outside code fences, is not reported as converged, the inventory includes tasks, and the docs describe it. Copilot's other note on #4330, that the dedup test did not pin substance-based matching, no longer applies, since that rule is gone.

Evidence

	on main	on this branch
test_checklist_scan_contract.py	3 failed, 2 passed	5 passed
test_taskstoissues_feature_scope.py	5 failed	5 passed
test_converge_prerequisite.py	7 failed	7 passed

• Each fix cherry-picked alone onto main: its own tests plus the goose integration tests pass (43, 43 and 45).
• Mutations, each reverted afterwards:
  • implement: removing the checked-items exclusion fails 1.
  • taskstoissues: restoring the previous legacy rule fails 2; removing the question to the user fails 1.
  • converge: restoring fix(converge): do not re-append work an unchecked task already tracks (#4269) #4330's template fails 6; dropping the fence exclusion from the gate fails 2 (this one is also caught by the checkbox-scan contract from section 1); removing STOP fails 1; removing the task inventory fails 1.
• uvx ruff@0.15.0 check src tests: clean. markdownlint reports the same number of findings on the four touched Markdown files as on main (all pre-existing, in untouched lines).
• tests/unit and tests/integrations on Windows: the only failures are symlink-privilege errors (WinError 1314); the two whose names do not say so fail identically on a clean checkout of main.

AI disclosure

Per CONTRIBUTING: this pull request (code, tests, measurements and this description) was developed with Claude Code as a coding agent.