fix: add mandatory test generation to quick-dev workflow

The quick-dev workflow never produced tests because no step required them.
This adds a mandatory Tests section to step-03 (implement) and step-oneshot,
adds test coverage verification to the step-04 acceptance auditor, and
strengthens the spec template to require test tasks in the plan.

Closes #2208

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

src/bmm-skills/4-implementation/bmad-quick-dev/spec-template.md

@@ -53,7 +53,7 @@ context: [] # optional: `{project-root}/`-prefixed paths to project-wide standar
 ## Tasks & Acceptance
 
 <!-- Tasks: backtick-quoted file path -- action -- rationale. Prefer one task per file; group tightly-coupled changes when splitting would be artificial. -->
-<!-- If an I/O Matrix is present, include a task to unit-test its edge cases. -->
+<!-- REQUIRED: Include test tasks for every new or modified behavior. If an I/O Matrix is present, include a task to unit-test its edge cases. -->
 <!-- AC covers system-level behaviors not captured by the I/O Matrix. Do not duplicate I/O scenarios here. -->
 
 **Execution:**

src/bmm-skills/4-implementation/bmad-quick-dev/step-03-implement.md

@@ -30,9 +30,24 @@ Hand `{spec_file}` to a sub-agent/task and let it implement. If no sub-agents ar
 
 **Path formatting rule:** Any markdown links written into `{spec_file}` must use paths relative to `{spec_file}`'s directory so they are clickable in VS Code. Any file paths displayed in terminal/conversation output must use CWD-relative format with `:line` notation (e.g., `src/path/file.ts:42`) for terminal clickability. No leading `/` in either case.
 
+### Tests
+
+**This is mandatory, not optional.** After implementation, write tests for every new or modified behavior. Follow the project's testing conventions discovered from `{project_context}`, CLAUDE.md, or the existing test suite.
+
+1. **Discover conventions.** Identify the project's test framework, file-naming patterns, and test directory structure from existing tests. If no existing tests exist, use the project's language-idiomatic defaults.
+2. **Write tests.** Cover:
+   - Happy-path behavior for each new or changed feature.
+   - Edge cases and error scenarios from the I/O & Edge-Case Matrix (if present in the spec).
+   - Regressions — any behavior that could break due to the change.
+3. **Run tests.** Execute the test suite. All new and existing tests must pass. If any test fails, fix the implementation or the test before proceeding.
+
+If the project has no test infrastructure at all (no test framework, no test directory, no test scripts), note this in the spec under `## Verification` and skip — but this is the only acceptable reason to skip tests.
+
 ### Self-Check
 
-Before leaving this step, verify every task in the `## Tasks & Acceptance` section of `{spec_file}` is complete. Mark each finished task `[x]`. If any task is not done, finish it before proceeding.
+Before leaving this step, verify:
+1. Every task in the `## Tasks & Acceptance` section of `{spec_file}` is complete. Mark each finished task `[x]`. If any task is not done, finish it before proceeding.
+2. Tests exist for every new or modified behavior (unless the project has no test infrastructure). If tests are missing, go back and write them before proceeding.
 
 ## NEXT
 

src/bmm-skills/4-implementation/bmad-quick-dev/step-04-review.md

@@ -26,7 +26,7 @@ Launch three subagents without conversation context. If no sub-agents are availa
 
 - **Blind hunter** — receives `{diff_output}` only. No spec, no context docs, no project access. Invoke via the `bmad-review-adversarial-general` skill.
 - **Edge case hunter** — receives `{diff_output}` and read access to the project. Invoke via the `bmad-review-edge-case-hunter` skill.
-- **Acceptance auditor** — receives `{diff_output}`, `{spec_file}`, and read access to the project. Must also read the docs listed in `{spec_file}` frontmatter `context`. Checks for violations of acceptance criteria, rules, and principles from the spec and context docs.
+- **Acceptance auditor** — receives `{diff_output}`, `{spec_file}`, and read access to the project. Must also read the docs listed in `{spec_file}` frontmatter `context`. Checks for violations of acceptance criteria, rules, and principles from the spec and context docs. **Must also verify that tests exist for every new or modified behavior** — missing test coverage for changed code is a `bad_spec` finding unless the project has no test infrastructure.
 
 ### Classify
 

src/bmm-skills/4-implementation/bmad-quick-dev/step-oneshot.md

@@ -16,6 +16,16 @@ spec_file: '' # set by step-01 before entering this step
 
 Implement the clarified intent directly.
 
+### Tests
+
+**This is mandatory, not optional.** After implementation, write tests for every new or modified behavior. Follow the project's testing conventions discovered from `{project_context}`, CLAUDE.md, or the existing test suite.
+
+1. **Discover conventions.** Identify the project's test framework, file-naming patterns, and test directory structure from existing tests. If no existing tests exist, use the project's language-idiomatic defaults.
+2. **Write tests.** Cover happy-path behavior, edge cases, and regressions.
+3. **Run tests.** All new and existing tests must pass. Fix failures before proceeding.
+
+If the project has no test infrastructure at all (no test framework, no test directory, no test scripts), skip — but this is the only acceptable reason to skip tests.
+
 ### Review
 
 Invoke the `bmad-review-adversarial-general` skill in a subagent with the changed files. The subagent gets NO conversation context — to avoid anchoring bias. If no sub-agents are available, write the changed files to a review prompt file in `{implementation_artifacts}` and HALT. Ask the human to run the review in a separate session and paste back the findings.