prizmkit 1.1.112 → 1.1.113

package/bundled/VERSION.json

@@ -1,5 +1,5 @@
 {
-  "frameworkVersion": "1.1.112",
-  "bundledAt": "2026-07-08T16:44:52.355Z",
-  "bundledFrom": "544bc17"
+  "frameworkVersion": "1.1.113",
+  "bundledAt": "2026-07-09T01:22:22.762Z",
+  "bundledFrom": "09ad715"
 }

package/bundled/adapters/codex/skill-adapter.js

@@ -21,6 +21,10 @@ function applyCodexInteractionCompatibility(content) {
 
   return content
     .replace(/\bAskUserQuestion\b/g, 'request_user_input')
+    .replace(
+      /Use `request_user_input`; do not render path options as plain text\./g,
+      fallbackInstruction
+    )
     .replace(
       /Do NOT render options as plain text — the user must be able to click\/select\./g,
       fallbackInstruction

package/bundled/dev-pipeline/SCHEMA_ANALYSIS.md

@@ -81,8 +81,8 @@ acceptance_criteria, status
 
 **Optional fields:**
 ```
-priority, affected_feature, affected_modules, environment, critic, 
-critic_count, model
+priority, dependencies, affected_feature, affected_modules, environment, critic, 
+critic_count, model, browser_interaction
 ```
 
 | Property | Type | Enum/Pattern | Notes |
@@ -95,7 +95,8 @@ critic_count, model
 | `error_source` | object | type enum: `stack_trace, user_report, failed_test, log_pattern, monitoring_alert` | ✓ REQUIRED |
 | `verification_type` | string | `automated, manual, hybrid` | ✓ REQUIRED |
 | `acceptance_criteria` | array | N/A | minItems: 1 |
-| `status` | string | `pending, triaging, reproducing, fixing, verifying, completed, failed, needs_info, skipped` | ✓ REQUIRED; more states than Feature |
+| `status` | string | `pending, in_progress, completed, failed, skipped, needs_info` | ✓ REQUIRED; external list status only |
+| `dependencies` | array | Pattern: `^B-\d{3}$` | Optional; bug IDs that must complete first |
 | `affected_feature` | string | Pattern: `^F-\d{3}(-[A-Z])?$` | Optional; link to original feature |
 | `affected_modules` | array | N/A | Optional; module names |
 | `environment` | object | N/A | Optional; os, runtime, browser, version |
@@ -209,12 +210,11 @@ design_system, ci_pipeline
 pending, in_progress, completed, failed, skipped, split, auto_skipped
 ```
 
-**Bug Fix Status (9 values):**
+**Bug Fix Status (6 external values):**
 ```
-pending, triaging, reproducing, fixing, verifying, completed, failed, needs_info, skipped
+pending, in_progress, completed, failed, skipped, needs_info
 ```
-*Includes intermediate workflow states: triaging, reproducing, fixing, verifying, needs_info*
-*NO "split" or "auto_skipped"*
+*External bug-list status stays coarse. Fine-grained triage/reproduction/fix/verification phases belong in logs, checkpoints, and session metadata, not `bugs[].status`.*
 
 **Refactor Status (5 values):**
 ```
@@ -227,10 +227,10 @@ pending, in_progress, completed, failed, skipped
 | Schema | Dependencies | Pattern |
 |--------|--------------|---------|
 | Feature | `dependencies` (required) | `^F-\d{3}(-[A-Z])?$` - can reference sub-features |
-| Bug Fix | N/A | No `dependencies` field ❌ |
+| Bug Fix | `dependencies` (optional) | `^B-\d{3}$` - can reference prerequisite bug fixes |
 | Refactor | `dependencies` (required) | `^R-\d{3}$` - ONLY refactor IDs, no sub-IDs |
 
-**Issue:** Bug Fix schema lacks dependency tracking. Bugs cannot declare prerequisites. Feature dependencies support sub-features; Refactor dependencies do not.
+**Contract:** Bug Fix dependencies are supported as optional prerequisites. Runtime selects eligible bugs in stable list order after dependencies are completed.
 
 ### 4.5 Acceptance Criteria vs Task Lists
 
@@ -493,7 +493,7 @@ Located in `/dev-pipeline/templates/`:
 
 ### Status Enums
 - **Feature**: `[pending, in_progress, completed, failed, skipped, split, auto_skipped]` (7 values)
-- **Bug Fix**: `[pending, triaging, reproducing, fixing, verifying, completed, failed, needs_info, skipped]` (9 values)
+- **Bug Fix**: `[pending, in_progress, completed, failed, skipped, needs_info]` (6 external values)
 - **Refactor**: `[pending, in_progress, completed, failed, skipped]` (5 values)
 
 ### Complexity/Estimated Complexity Enums

package/bundled/dev-pipeline/scripts/generate-bootstrap-prompt.py

@@ -1906,6 +1906,7 @@ def build_replacements(args, feature, features, global_context, script_dir):
         "{{HAS_PLAN}}": "true" if artifacts["has_plan"] else "false",
         "{{BROWSER_VERIFY_STEPS}}": browser_verify_steps,
         "{{BROWSER_TOOL}}": browser_tool,
+        "{{BROWSER_BLOCKING}}": "true" if browser_blocking else "false",
         "{{AC_CHECKLIST}}": format_ac_checklist(
             feature.get("acceptance_criteria", [])
         ),

package/bundled/dev-pipeline/scripts/generate-bugfix-prompt.py

@@ -417,6 +417,13 @@ def assemble_bugfix_sections(pipeline_mode, sections_dir, critic_enabled, browse
         sections.append(("phase-browser", load_section(sections_dir, "phase-browser-verification-auto.md")))
     if manual_gate:
         sections.append(("bugfix-manual-gate", load_section(sections_dir, "bugfix-phase-manual-verification.md")))
+        sections.extend([
+            ("bugfix-session-status", load_section(sections_dir, "bugfix-session-status.md")),
+            ("bugfix-critical-paths", load_section(sections_dir, "bugfix-critical-paths.md")),
+            ("failure-capture", load_section(sections_dir, "failure-capture.md")),
+            ("bugfix-reminders", load_section(sections_dir, "bugfix-reminders.md")),
+        ])
+        return sections
     sections.extend([
         ("bugfix-phase-commit-report", load_section(sections_dir, "bugfix-phase-commit-report.md")),
         ("bugfix-session-status", load_section(sections_dir, "bugfix-session-status.md")),

package/bundled/dev-pipeline/templates/bootstrap-tier1.md

@@ -338,7 +338,13 @@ Append results to `context-snapshot.md`:
    Browser cleanup: confirmed
    ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 {{END_IF_BROWSER_INTERACTION}}
 
 ### Phase 4: Architecture Sync & Commit (SINGLE COMMIT)

package/bundled/dev-pipeline/templates/bootstrap-tier2.md

@@ -488,7 +488,13 @@ Append results to `context-snapshot.md`:
    Browser cleanup: confirmed
    ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 {{END_IF_BROWSER_INTERACTION}}
 
 ### Phase 6: Architecture Sync & Commit (SINGLE COMMIT)

package/bundled/dev-pipeline/templates/bootstrap-tier3.md

@@ -483,7 +483,13 @@ Append results to `context-snapshot.md`:
    Browser cleanup: confirmed
    ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 {{END_IF_BROWSER_INTERACTION}}
 
 ### Phase 6: Retrospective & Commit (SINGLE COMMIT) — DO NOT SKIP

package/bundled/dev-pipeline/templates/sections/bugfix-phase-commit-report.md

@@ -1,6 +1,6 @@
 ### Phase 4: Commit & Learn
 
-Skip this phase only when the Manual / Hybrid Verification Gate stopped the fresh session before commit.
+This phase is included only for sessions allowed to commit. Fresh manual/hybrid bugfix prompts omit this phase entirely and stop as partial before retrospective/commit.
 
 **Bug Fix Documentation Policy**:
 - Default: run `/prizmkit-retrospective` with structural sync only.

package/bundled/dev-pipeline/templates/sections/phase-browser-verification-auto.md

@@ -117,5 +117,11 @@ Append results to `context-snapshot.md`:
    Browser cleanup: confirmed
    ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 

package/bundled/dev-pipeline/templates/sections/phase-browser-verification-opencli.md

@@ -114,5 +114,11 @@ Append results to `context-snapshot.md`:
    Browser cleanup: confirmed
    ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 

package/bundled/dev-pipeline/templates/sections/phase-browser-verification.md

@@ -101,5 +101,11 @@ Server cleanup: confirmed
 Browser cleanup: confirmed
 ```
 
-If verification fails, log the failure details but continue to commit. Failures do NOT block the commit, but you MUST attempt verification and MUST clean up the dev server.
+Browser blocking policy: `{{BROWSER_BLOCKING}}`.
+
+If verification fails:
+- If `{{BROWSER_BLOCKING}}` is `true`: mark checkpoint step `browser-verification` as `failed`, write `failure-log.md` with browser evidence, clean up the dev server/browser, and stop before commit.
+- If `{{BROWSER_BLOCKING}}` is `false`: log the failure details, mark checkpoint step `browser-verification` according to the evidence, clean up the dev server/browser, and continue.
+
+You MUST attempt verification and MUST clean up the dev server in both cases.
 

package/bundled/dev-pipeline/tests/test_generate_bootstrap_prompt.py

@@ -1100,7 +1100,10 @@ def _ensure_agent_files(project, platforms=("claude", "codex")):
             (agents_dir / f"prizm-dev-team-{name}{suffix}").write_text("agent\n", encoding="utf-8")
 
 
-def _render_bugfix_prompt(tmp_path, platform="claude", explicit_platform=None):
+def _render_bugfix_prompt(
+    tmp_path, platform="claude", explicit_platform=None,
+    bug_overrides=None, retry_count="0", resume_phase="null",
+):
     from generate_bugfix_prompt import main as bugfix_main
 
     project = tmp_path / "bugfix-project"
@@ -1112,8 +1115,11 @@ def _render_bugfix_prompt(tmp_path, platform="claude", explicit_platform=None):
     _ensure_agent_files(project)
     bug_list = project / ".prizmkit" / "plans" / "bug-fix-list.json"
     bug_list.parent.mkdir(parents=True)
+    bug = {"id": "B-001", "title": "Broken", "description": "Fix it", "acceptance_criteria": ["Works"]}
+    if bug_overrides:
+        bug.update(bug_overrides)
     bug_list.write_text(json.dumps({
-        "bugs": [{"id": "B-001", "title": "Broken", "description": "Fix it", "acceptance_criteria": ["Works"]}],
+        "bugs": [bug],
         "global_context": {"language": "Python"},
     }), encoding="utf-8")
     output = project / "bugfix-prompt.md"
@@ -1130,8 +1136,8 @@ def _render_bugfix_prompt(tmp_path, platform="claude", explicit_platform=None):
             "--bug-id", "B-001",
             "--session-id", "session-1",
             "--run-id", "run-1",
-            "--retry-count", "0",
-            "--resume-phase", "null",
+            "--retry-count", retry_count,
+            "--resume-phase", resume_phase,
             "--state-dir", str(project / ".prizmkit" / "state" / "bugfix"),
             "--project-root", str(project),
             "--output", str(output),
@@ -1150,7 +1156,7 @@ def _render_bugfix_prompt(tmp_path, platform="claude", explicit_platform=None):
     return output.read_text(encoding="utf-8")
 
 
-def _render_refactor_prompt(tmp_path, platform="claude", explicit_platform=None):
+def _render_refactor_prompt(tmp_path, platform="claude", explicit_platform=None, refactor_overrides=None):
     from generate_refactor_prompt import main as refactor_main
 
     project = tmp_path / "refactor-project"
@@ -1162,14 +1168,17 @@ def _render_refactor_prompt(tmp_path, platform="claude", explicit_platform=None)
     _ensure_agent_files(project)
     refactor_list = project / ".prizmkit" / "plans" / "refactor-list.json"
     refactor_list.parent.mkdir(parents=True)
+    refactor = {
+        "id": "R-001",
+        "title": "Restructure",
+        "description": "Preserve behavior",
+        "acceptance_criteria": ["Behavior preserved"],
+        "scope": {"files": ["src/a.py"], "modules": ["src"]},
+    }
+    if refactor_overrides:
+        refactor.update(refactor_overrides)
     refactor_list.write_text(json.dumps({
-        "refactors": [{
-            "id": "R-001",
-            "title": "Restructure",
-            "description": "Preserve behavior",
-            "acceptance_criteria": ["Behavior preserved"],
-            "scope": {"files": ["src/a.py"], "modules": ["src"]},
-        }],
+        "refactors": [refactor],
         "global_context": {"language": "Python"},
     }), encoding="utf-8")
     output = project / "refactor-prompt.md"
@@ -1189,18 +1198,14 @@ def _render_refactor_prompt(tmp_path, platform="claude", explicit_platform=None)
             "--retry-count", "0",
             "--resume-phase", "null",
             "--state-dir", str(project / ".prizmkit" / "state" / "refactor"),
+            "--project-root", str(project),
             "--output", str(output),
         ]
-        old_cwd = os.getcwd()
-        os.chdir(project)
         try:
-            try:
-                refactor_main()
-            except SystemExit as exc:
-                if exc.code not in (0, None):
-                    raise
-        finally:
-            os.chdir(old_cwd)
+            refactor_main()
+        except SystemExit as exc:
+            if exc.code not in (0, None):
+                raise
     finally:
         os.sys.argv = old_argv
         if old_platform is None:
@@ -1210,6 +1215,67 @@ def _render_refactor_prompt(tmp_path, platform="claude", explicit_platform=None)
     return output.read_text(encoding="utf-8")
 
 
+class TestBrowserBlockingPromptPolicy:
+    def test_feature_bugfix_refactor_blocking_browser_policy_stops_before_commit(self, tmp_path):
+        browser_interaction = {
+            "tool": "auto",
+            "blocking": True,
+            "verify_steps": ["Open the page"],
+        }
+        prompts = [
+            _render_feature_prompt(tmp_path / "feature-browser-blocking", browser_interaction=browser_interaction),
+            _render_bugfix_prompt(tmp_path / "bugfix-browser-blocking", bug_overrides={"browser_interaction": browser_interaction}),
+            _render_refactor_prompt(tmp_path / "refactor-browser-blocking", refactor_overrides={"browser_interaction": browser_interaction}),
+        ]
+
+        for prompt in prompts:
+            assert "Browser blocking policy: `true`" in prompt
+            assert "stop before commit" in prompt
+            assert "Failures do NOT block the commit" not in prompt
+
+    def test_feature_bugfix_refactor_nonblocking_browser_policy_continues(self, tmp_path):
+        browser_interaction = {
+            "tool": "auto",
+            "blocking": False,
+            "verify_steps": ["Open the page"],
+        }
+        prompts = [
+            _render_feature_prompt(tmp_path / "feature-browser-nonblocking", browser_interaction=browser_interaction),
+            _render_bugfix_prompt(tmp_path / "bugfix-browser-nonblocking", bug_overrides={"browser_interaction": browser_interaction}),
+            _render_refactor_prompt(tmp_path / "refactor-browser-nonblocking", refactor_overrides={"browser_interaction": browser_interaction}),
+        ]
+
+        for prompt in prompts:
+            assert "Browser blocking policy: `false`" in prompt
+            assert "clean up the dev server/browser, and continue" in prompt
+            assert "Failures do NOT block the commit" not in prompt
+
+
+class TestManualHybridBugfixPromptPolicy:
+    def test_fresh_manual_bugfix_prompt_stops_before_commit_phase(self, tmp_path):
+        prompt = _render_bugfix_prompt(
+            tmp_path / "fresh-manual-bugfix",
+            bug_overrides={"verification_type": "manual"},
+        )
+
+        assert "Manual / Hybrid Verification Gate" in prompt
+        assert "Stop before `/prizmkit-retrospective` and `/prizmkit-committer`" in prompt
+        assert "### Phase 4: Commit & Learn" not in prompt
+        assert "Run `/prizmkit-committer` with commit prefix" not in prompt
+
+    def test_recovery_manual_bugfix_prompt_includes_commit_phase_with_substitution_policy(self, tmp_path):
+        prompt = _render_bugfix_prompt(
+            tmp_path / "recovery-manual-bugfix",
+            bug_overrides={"verification_type": "manual"},
+            retry_count="1",
+        )
+
+        assert "Manual / Hybrid Verification Gate" not in prompt
+        assert "Recovery-class session: automated verification may substitute for manual/hybrid UAT" in prompt
+        assert "### Phase 4: Commit & Learn" in prompt
+        assert "Run `/prizmkit-committer` with commit prefix" in prompt
+
+
 class TestHeadlessPromptCleanupF033:
     def test_rendered_feature_bugfix_refactor_prompts_are_clean(self, tmp_path):
         prompts = []

package/bundled/dev-pipeline/tests/test_generate_bugfix_prompt.py

@@ -307,6 +307,20 @@ def test_template_source_is_clean_for_headless_prompt_contract():
         assert forbidden not in content
 
 
+def test_bugfix_prompt_keeps_fine_grained_phases_out_of_external_status_contract():
+    content = Path("dev-pipeline/templates/bugfix-bootstrap-prompt.md").read_text(encoding="utf-8")
+    old_statuses = ["triaging", "reproducing", "fixing", "verifying"]
+
+    assert "external bug-list status field" in content
+    assert "pending`, `in_progress`, `completed`, `failed`, `skipped`, or `needs_info`" in content
+    for status in old_statuses:
+        assert f"status=\"{status}\"" not in content
+        assert f"status = \"{status}\"" not in content
+        assert f"status='{status}'" not in content
+        assert f"status = '{status}'" not in content
+        assert f"Set bug status to `{status}`" not in content
+
+
 # ---------------------------------------------------------------------------
 # Anti-Bash regression guardrails (F-039)
 # Note: grep -n, sed -n, grep -q, wc -l excluded — referenced in bugfix context

package/bundled/dev-pipeline/tests/test_python_runner_parity.py

@@ -514,6 +514,48 @@ def test_init_scripts_reject_missing_self_and_cycle_dependencies(tmp_path):
             assert any(expected_fragments[scenario] in error for error in payload["errors"])
 
 
+def test_bug_planner_validator_rejects_dependency_cycles():
+    from importlib.util import module_from_spec, spec_from_file_location
+
+    script_path = REPO_ROOT / "core" / "skills" / "orchestration-skill" / "pipelines" / "bug-planner" / "scripts" / "validate-bug-list.py"
+    spec = spec_from_file_location("bug_planner_validate_bug_list", script_path)
+    module = module_from_spec(spec)
+    spec.loader.exec_module(module)
+
+    result = module.validate({
+        "$schema": "dev-pipeline-bug-fix-list-v1",
+        "project_name": "Fake",
+        "bugs": [
+            {
+                "id": "B-001",
+                "title": "First",
+                "description": "First bug",
+                "severity": "medium",
+                "error_source": {"type": "user_report"},
+                "verification_type": "manual",
+                "acceptance_criteria": ["fixed"],
+                "dependencies": ["B-002"],
+                "status": "pending",
+            },
+            {
+                "id": "B-002",
+                "title": "Second",
+                "description": "Second bug",
+                "severity": "medium",
+                "error_source": {"type": "user_report"},
+                "verification_type": "manual",
+                "acceptance_criteria": ["fixed"],
+                "dependencies": ["B-001"],
+                "status": "pending",
+            },
+        ],
+    })
+
+    assert result["valid"] is False
+    assert any("Dependency cycle detected" in error for error in result["errors"])
+
+
+
 def _family_items(kind, entries):
     items = []
     for item_id, dependencies, status in entries:

package/bundled/skills/_metadata.json

@@ -1,5 +1,5 @@
 {
-  "version": "1.1.112",
+  "version": "1.1.113",
   "skills": {
     "prizmkit": {
       "description": "Full-lifecycle dev toolkit. Covers spec-driven development, Prizm context docs, code quality, debugging, deployment, and knowledge management.",

package/bundled/skills/bug-planner/scripts/validate-bug-list.py

@@ -34,6 +34,38 @@ BUG_ID_PATTERN = re.compile(r"^B-\d{3}$")
 SCHEMA_VERSION = "dev-pipeline-bug-fix-list-v1"
 
 
+def _find_dependency_cycle(dep_map, item_ids):
+    """Return one dependency cycle as a list of bug IDs, or an empty list."""
+    visiting = set()
+    visited = set()
+    stack = []
+
+    def visit(node):
+        visiting.add(node)
+        stack.append(node)
+        for dep in dep_map.get(node, []):
+            if dep not in item_ids:
+                continue
+            if dep in visiting:
+                start = stack.index(dep)
+                return stack[start:] + [dep]
+            if dep not in visited:
+                cycle = visit(dep)
+                if cycle:
+                    return cycle
+        stack.pop()
+        visiting.remove(node)
+        visited.add(node)
+        return []
+
+    for bug_id in sorted(item_ids):
+        if bug_id not in visited:
+            cycle = visit(bug_id)
+            if cycle:
+                return cycle
+    return []
+
+
 def _err(msg):
     """Print an error message to stderr."""
     print("ERROR: {}".format(msg), file=sys.stderr)
@@ -181,15 +213,23 @@ def validate(data, feature_ids=None):
         for bug in bugs
         if BUG_ID_PATTERN.match(str(bug.get("id", "")))
     }
+    dep_map = {bug_id: [] for bug_id in known_ids}
     for i, bug in enumerate(bugs):
         bug_id = bug.get("id", "")
         deps = bug.get("dependencies", []) or []
         if not isinstance(deps, list):
             continue
         for dep in deps:
-            if isinstance(dep, str) and BUG_ID_PATTERN.match(dep) and dep not in known_ids:
-                errors.append("bugs[{}] ({}): dependency '{}' does not exist".format(
-                    i, bug_id, dep))
+            if isinstance(dep, str) and BUG_ID_PATTERN.match(dep):
+                if dep not in known_ids:
+                    errors.append("bugs[{}] ({}): dependency '{}' does not exist".format(
+                        i, bug_id, dep))
+                elif BUG_ID_PATTERN.match(str(bug_id)):
+                    dep_map.setdefault(bug_id, []).append(dep)
+
+    cycle = _find_dependency_cycle(dep_map, known_ids)
+    if cycle:
+        errors.append("Dependency cycle detected: {}".format(" -> ".join(cycle)))
 
     return {
         "valid": len(errors) == 0,

package/bundled/skills/bugfix-pipeline-launcher/SKILL.md

@@ -5,7 +5,7 @@ description: "Launch and manage the bugfix pipeline from within an AI CLI sessio
 
 # Bugfix-Pipeline Launcher
 
-Launch the autonomous bug fix pipeline from within an AI CLI conversation. Supports foreground and background execution modes.
+Launch the autonomous bug fix pipeline from within an AI CLI conversation. Only Background daemon mode is fully detached and survives AI CLI session closure; Foreground runs in the current session, and Manual prints commands without launching.
 
 ### Execution Mode
 
@@ -15,7 +15,7 @@ Three execution modes are available. The user chooses one before configuring oth
 2. **Background daemon** — `python3 ./.prizmkit/dev-pipeline/cli.py daemon bugfix start .prizmkit/plans/bug-fix-list.json`. Runs fully detached, survives AI CLI session closure.
 3. **Manual** — Display the assembled command(s) only. Do not execute anything. User runs them on their own.
 
-**Background mode traceability**: Daemon mode metadata and logs are written by the Python runtime (`daemon bugfix start/status/logs`). Do not append a separate skill-managed audit log.
+**Background mode traceability**: Daemon mode metadata and logs are written by the Python runtime (`daemon bugfix start/status/logs`). Do not create any extra launcher-managed audit file.
 
 ### When to Use
 
@@ -90,7 +90,6 @@ Detect user intent from their message, then follow the corresponding workflow:
        data = json.load(f)
    bugs = data.get('bugs', [])
    severity_order = {'critical': 0, 'high': 1, 'medium': 2, 'low': 3}
-   bugs_sorted = sorted(bugs, key=lambda b: (severity_order.get(b.get('severity', 'medium'), 2), b.get('priority', 99)))
    print(f'Total bugs: {len(bugs)}')
    sev_counts = {}
    for b in bugs:
@@ -98,7 +97,7 @@ Detect user intent from their message, then follow the corresponding workflow:
        sev_counts[s] = sev_counts.get(s, 0) + 1
    print(f'By severity: {dict(sorted(sev_counts.items(), key=lambda x: severity_order.get(x[0], 2)))}')
    print()
-   for b in bugs_sorted:
+   for b in bugs:
        print(f\"  {b['id']}: [{b.get('severity','?').upper()}] {b.get('title', 'untitled')}\")
    "
    ```
@@ -165,6 +164,7 @@ Detect user intent from their message, then follow the corresponding workflow:
    - Summarize results: total bugs, fixed, failed, skipped
    - If all fixed: each bug session has already run `prizmkit-retrospective` internally (structural sync by default; full retrospective when the fix changed interfaces, dependencies, or observable behavior). Ask user what's next.
    - If some failed: show failed bug IDs and suggest `python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix <B-XXX> --clean --run` for a fresh retry
+   - **Browser verification**: Bug sessions with `browser_interaction` perform their own verification attempts during execution. After completion, use `python3 ./.prizmkit/dev-pipeline/cli.py bugfix status .prizmkit/plans/bug-fix-list.json` and per-bug session logs as the source of truth before summarizing browser verification results.
 
    **If background daemon**:
    1. Verify launch:
@@ -264,9 +264,9 @@ Read `${SKILL_DIR}/references/configuration.md` for the full error handling tabl
 ### Integration Notes
 
 - **After bug-planner**: This is the natural next step. When user finishes bug planning and has `.prizmkit/plans/bug-fix-list.json`, suggest launching the bugfix pipeline.
-- **Session independence**: In daemon mode, the bugfix pipeline runs completely detached. User can close the AI CLI, open a new session later, and use this skill to check progress or stop the pipeline.
+- **Session independence**: Only Background daemon mode runs detached. Foreground runs in the current AI CLI session, and Manual mode prints commands without launching.
 - **Single instance per family**: Only one bugfix pipeline can run at a time. Different pipeline families may coexist because they use separate daemon metadata and state directories.
-- **Feature pipeline coexistence**: Bugfix and feature pipelines use separate state directories (`.prizmkit/state/bugfix/` vs `.prizmkit/state/features/`), so they can run simultaneously without conflict.
+- **Pipeline coexistence**: Bugfix, feature, and refactor pipelines use separate state directories (`.prizmkit/state/bugfix/`, `.prizmkit/state/features/`, `.prizmkit/state/refactor/`), so all three can run simultaneously without conflict.
 - **State preservation**: Stopping and restarting the bugfix pipeline resumes from where it left off -- completed bugs are not re-fixed.
 - **Bug ordering**: Dependencies are active. Runtime selects eligible bugs in stable list order after dependencies are completed; planners should order unrelated bugs by severity/priority before launch.
 - **Background mode traceability**: Daemon mode metadata/logging is runtime-owned; use `daemon bugfix status` and `daemon bugfix logs`.

package/bundled/skills/feature-pipeline-launcher/SKILL.md

@@ -309,22 +309,13 @@ Notes:
 
 After pipeline completion, if features have `browser_interaction` fields and the corresponding browser tool (`playwright-cli` or `opencli`) is installed:
 
-If list status may be stale, prefer the runtime status command before deciding which completed features qualify:
+If list status may be stale, use the runtime status command as the source of truth for completion state:
 `python3 ./.prizmkit/dev-pipeline/cli.py feature status .prizmkit/plans/feature-list.json`.
 
-1. **Check which features qualify**:
-   ```bash
-   python3 -c "
-   import json
-   with open('.prizmkit/plans/feature-list.json') as f:
-       data = json.load(f)
-   for feat in data.get('features', []):
-       bi = feat.get('browser_interaction')
-       if bi and feat.get('status') == 'completed':
-           tool = bi.get('tool', 'auto') if isinstance(bi, dict) else 'auto'
-           print(f\"  {feat['id']}: {feat.get('title','')} (tool: {tool})\")
-   " 2>/dev/null
-   ```
+1. **Check which completed features have browser verification configured**:
+   - Read `browser_interaction` configuration from `.prizmkit/plans/feature-list.json`.
+   - Use the runtime status command output to decide which configured features are actually completed.
+   - Do not rely on list-file `status` alone when runtime state exists, because daemon/foreground bookkeeping may be newer than the plan file.
 
 2. **Ask user**: "N features have browser verification configured. Run browser verification now? (Y/n)"
 
@@ -352,8 +343,8 @@ Read `${SKILL_DIR}/references/configuration.md` for the full error handling tabl
 ### Integration Notes
 
 - **After feature-planner**: This is the natural next step. When user finishes planning and has `.prizmkit/plans/feature-list.json`, suggest launching the pipeline.
-- **Session independence**: The pipeline runs completely detached. User can close the AI CLI session, open a new session later, and use this skill to check progress or stop the pipeline.
+- **Session independence**: Only Background daemon mode runs detached. Foreground runs in the current AI CLI session, and Manual mode prints commands without launching.
 - **Single instance per family**: Only one feature pipeline can run at a time. Different pipeline families may coexist because they use separate daemon metadata and state directories.
-- **Pipeline coexistence**: Feature and bugfix pipelines use separate state directories (`.prizmkit/state/features/` vs `.prizmkit/state/bugfix/`), so they can run simultaneously without conflict.
+- **Pipeline coexistence**: Feature, bugfix, and refactor pipelines use separate state directories (`.prizmkit/state/features/`, `.prizmkit/state/bugfix/`, `.prizmkit/state/refactor/`), so all three can run simultaneously without conflict.
 - **State preservation**: Stopping and restarting the pipeline resumes from where it left off -- completed features are not re-run.
 - **HANDOFF**: After pipeline completes all features, each session has already run `prizmkit-retrospective` internally. Ask user what's next.

package/bundled/skills/feature-workflow/SKILL.md

@@ -87,7 +87,7 @@ For existing-project feature work:
 Question: How would you like to proceed?
 Header: Approach
 Options:
-- Implement now (Fast Path) — current session/current workspace via /prizmkit-plan -> /prizmkit-implement -> review/retro/commit
+- Implement now (Fast Path) — current session/current workspace via simplified /prizmkit-plan -> /prizmkit-implement -> conditional test/review/retro -> /prizmkit-committer
 - Use Pipeline Path — feature-planner creates/reviews feature-list.json, then feature-pipeline-launcher configures execution
 - Continue clarifying — ask a few more route-level questions before choosing
 ```
@@ -166,12 +166,13 @@ Do not generate `feature-list.json` inside this workflow; `feature-planner` owns
 
 Use Fast Path only after explicit user selection.
 
-1. Invoke `/prizmkit-plan` with the handoff brief.
+1. Invoke `/prizmkit-plan` with the handoff brief; use simplified planning for this scoped feature.
 2. Invoke `/prizmkit-implement` after the plan is ready.
-3. Run `/prizmkit-code-review` as the quality gate.
-4. Run `/prizmkit-retrospective` for feature changes before commit.
-5. Run `/prizmkit-committer` with a `feat(<scope>):` message.
-6. End the workflow after commit; do not call `feature-planner`, `feature-pipeline-launcher`, or monitoring.
+3. Run `/prizmkit-test` only when risk triggers match: behavior, public interface, data model, security/permission/payment, deployment readiness, or explicit user verification request.
+4. Run `/prizmkit-code-review` when the user requested review, implementation uncertainty is high, or Full-path-level risk appears during implementation.
+5. Run `/prizmkit-retrospective` only when structure, interfaces, dependencies, behavior, or durable project knowledge changed.
+6. Run `/prizmkit-committer` with a `feat(<scope>):` message.
+7. End the workflow after commit; do not call `feature-planner`, `feature-pipeline-launcher`, or monitoring.
 
 If implementation scope grows beyond the Fast Path criteria, pause and ask whether to switch to Pipeline Path.
 
@@ -236,6 +237,6 @@ For start/stop/logs/retry execution, hand off to `feature-pipeline-launcher`; do
 
 ## Output
 
-- Fast Path: `spec.md`, `plan.md`, code changes, review result, Prizm docs sync, and one feature commit.
+- Fast Path: concise `spec.md`, simplified `plan.md`, code changes, conditional test/review evidence when triggered, Prizm docs sync only when durable knowledge changed, and one feature commit.
 - Pipeline Path: handoff brief, validated `.prizmkit/plans/feature-list.json` from `feature-planner`, and pipeline launch/status handled by `feature-pipeline-launcher`.
 - Greenfield path: app-level planning artifacts from `app-planner`, optionally followed by feature planning and launch.

package/bundled/skills/prizmkit-code-review/references/reviewer-agent-prompt.md

@@ -12,7 +12,7 @@ You are the dedicated PrizmKit Reviewer agent (`prizm-dev-team-reviewer`). Revie
 {architecture decisions and task list from plan.md}
 
 ## Scoped Test Report
-{summary from artifact_dir/test-report-path.txt and test-report.md, including Scope, Generated / Updated Tests, In-Scope Failures, Baseline Failures, Out of Scope Gaps, Boundary Matrix / Boundary Completion Gate if present, and Verdict. If no scoped report exists for a feature artifact, write "No scoped /prizmkit-test report found (expected for pre-test review)." Treat that absence as informational unless the calling workflow explicitly requires a prior test report.}
+{summary from artifact_dir/test-report-path.txt and test-report.md, including Scope, Generated / Updated Tests, In-Scope Failures, Baseline Failures, Out of Scope Gaps, Boundary Matrix / Boundary Completion Gate if present, and Verdict. If no scoped report exists for a change artifact, write "No scoped /prizmkit-test report found (expected for pre-test review)." Treat that absence as informational unless the calling workflow explicitly requires a prior test report.}
 
 ## Workspace Context
 {git status, unstaged diff summary, staged diff summary, current working directory, expected active checkout git top-level, and new-file paths captured by the Main Agent in Phase 0 from the active checkout. This is your primary review scope — use this context to understand what changed. Request targeted file reads only when you need specific line-level detail for a finding, not to re-discover or re-index the repository. Do not read files that are not in the workspace context unless you need them to verify impact (callers, dependents, shared patterns). Do not switch to a separate worktree or repository copy — the Main Agent has provided everything needed to understand the change set.}
@@ -49,7 +49,7 @@ Evaluate the changes across these dimensions (focus on what's relevant):
 3. **Completeness**: Files that should have been changed but weren't? Missing tests, types, imports, exports?
 4. **Consistency**: Do changes follow the project's existing patterns, naming conventions, and code style?
 5. **Security**: Hardcoded secrets, injection vulnerabilities, unsafe operations.
-6. **Test quality**: If a scoped `/prizmkit-test` report is present, verify the report scope is `this-change`, the artifact dir matches the feature, generated/updated tests assert real behavior rather than mock success or empty assertions, in-scope failures are not mislabeled as baseline failures, baseline failures are genuinely unrelated or pre-existing, out-of-scope gaps were not turned into unrelated test changes, and the report verdict is compatible with proceeding.
+6. **Test quality**: If a scoped `/prizmkit-test` report is present, verify the report scope is `this-change`, the artifact dir matches the reviewed change, generated/updated tests assert real behavior rather than mock success or empty assertions, in-scope failures are not mislabeled as baseline failures, baseline failures are genuinely unrelated or pre-existing, out-of-scope gaps were not turned into unrelated test changes, and the report verdict is compatible with proceeding.
 7. **Rules compliance**: (Skip this dimension if no dev rules were provided.) Do changes follow the per-layer dev rules? Flag violations of framework conventions, naming patterns, state management, or other rules defined for that layer.
 
 ## Output Format

package/bundled/skills/prizmkit-deploy/references/database-setup.md

@@ -10,7 +10,14 @@ During Discovery Step 1 (Project Detection), database drivers were already scann
 > - **Yes** → proceed with database installation
 > - **No** → skip, record in deploy.md with note "database must be configured by user"
 
-## PostgreSQL setup
+## Secrets file safety
+
+Before writing `.prizmkit/deploy/secrets.local.json`:
+
+1. Verify `.gitignore` or an equivalent ignore rule covers `.prizmkit/deploy/secrets.local.json` and any `.prizmkit/deploy/*.local.json` pattern used by the project.
+2. If the ignore rule is missing, pause and add or request the ignore rule before writing secrets.
+3. Set the secrets file permissions to current-user read/write only, for example `chmod 600 .prizmkit/deploy/secrets.local.json` on POSIX systems.
+4. Record only presence metadata in deploy config, docs, reports, and history. Never copy raw passwords, connection strings, passphrases, decryption keys, unsalted hashes, or full env var values into committed files.
 
 ```
 apt-get install -y postgresql postgresql-contrib
@@ -21,7 +28,8 @@ sudo -u postgres psql -c "GRANT ALL ON DATABASE <project> TO <project>;"
 
 - Generate a secure random password (32 chars, alphanumeric + symbols).
 - Write the connection string to `.prizmkit/deploy/secrets.local.json`: `DATABASE_URL=postgresql://<project>:<password>@localhost:5432/<project>`.
-- In deploy.md, write: "PostgreSQL installed, connection info recorded in `.prizmkit/deploy/secrets.local.json` (not committed to git)".
+- Set `.prizmkit/deploy/secrets.local.json` permissions to current-user read/write only after writing.
+- In deploy.md, write: "PostgreSQL installed, connection info recorded in `.prizmkit/deploy/secrets.local.json` (gitignored; raw value not committed)".
 
 ## MySQL setup (future)
 
@@ -37,10 +45,11 @@ redis-cli CONFIG REWRITE
 
 - Bind to localhost only (modify `/etc/redis/redis.conf` if needed).
 - Write `REDIS_URL=redis://:<password>@localhost:6379` to `.prizmkit/deploy/secrets.local.json`.
+- Set `.prizmkit/deploy/secrets.local.json` permissions to current-user read/write only after writing.
 
 ## Security notes
 
 - Never write database passwords to deploy.md, because deploy.md may be committed to git and passwords would leak.
-- Passwords stored in `.prizmkit/deploy/secrets.local.json` (gitignored).
+- Passwords stored in `.prizmkit/deploy/secrets.local.json` only after verifying the file is ignored by git and locked down to current-user read/write permissions.
 - Default: database binds to localhost, no external access, because most indie projects only need local connections.
 - Record a `"database-setup"` event in deploy history (presence metadata only, no passwords).

package/bundled/skills/prizmkit-prizm-docs/references/op-init.md

@@ -21,8 +21,8 @@ STEPS:
    - Top-level module M → write .prizmkit/prizm-docs/<M>.prizm (include SUBDIRS section with pointers to sub-module docs)
    - Sub-module S inside M → write .prizmkit/prizm-docs/<M>/<S>.prizm
    Each L1 includes MODULE (full relative path), FILES count, RESPONSIBILITY, SUBDIRS with pointers, KEY_FILES (5-10 most important), DEPENDENCIES (imports, imported-by, external), RULES (1-3 most critical only). L1 does NOT include INTERFACES, DATA_FLOW, TRAPS, or DECISIONS (those are L2, generated lazily). Max 4KB each.
-6. Skip L2 docs during init — L2 is created lazily on first file modification or when AI needs deep understanding (ON_DEEP_READ trigger). L2 contains behavioral detail: INTERFACES, DATA_FLOW, TRAPS, DECISIONS, full RULES, domain-specific sections.
-7. Configure UserPromptSubmit hook in platform settings per ${SKILL_DIR}/assets/prizm-docs-format.md.md Section 11.
+6. Skip L2 docs during init — L2 docs are created after real context exists, not as placeholders. During implementation, if relevant L2 is missing, the implementing skill reads target source files as fallback and proceeds. `/prizmkit-retrospective` creates or updates L2 after the change only when durable structure, behavior, dependencies, interfaces, or TRAPS exist. Explicit `/prizmkit-prizm-docs` repair/rebuild may also create L2 when fixing out-of-band doc drift.
+7. Configure UserPromptSubmit hook in platform settings per ${SKILL_DIR}/assets/prizm-docs-format.md Section 11.
 8. Validate all generated docs: size limits (L0 <= 4KB, L1 <= 4KB), pointer resolution (every -> reference resolves), no circular dependencies, KEY: value format compliance, no anti-patterns (prose, code blocks, markdown headers), L1 does not contain INTERFACES/DATA_FLOW/TRAPS/DECISIONS, no CHANGELOG sections/files, no UPDATED/date metadata, no feature/bug/refactor/task/session/run/pipeline/workflow IDs, no branch names, no absolute worktree paths, and no `.prizmkit/specs` / `.prizmkit/dev-pipeline` artifact paths.
 9. Report summary: modules discovered, L1 docs generated, files excluded, warnings.
 
@@ -30,16 +30,14 @@ OUTPUT: List of generated files, module count, and validation results.
 
 ## Post-Init Behavior
 
-After initialization, L2 docs are created incrementally:
+After initialization, L2 docs are created only when durable context exists:
 
-ON_MODIFY trigger:
-- First time AI modifies a file in sub-module S within module M:
-  IF .prizmkit/prizm-docs/<M>/<S>.prizm does not exist:
-    AI reads the source files in S, generates L2 doc, then proceeds with modification
-- This ensures L2 docs have real depth, written when AI has actual context
+ON_MODIFY behavior:
+- When an AI modifies a file in sub-module S within module M and `.prizmkit/prizm-docs/<M>/<S>.prizm` does not exist, the implementation reads the source files in S as fallback context and proceeds.
+- Do not block implementation only to create a placeholder L2 doc.
+- After the change, run `/prizmkit-retrospective` when structure, interfaces, dependencies, behavior, or durable TRAPS changed; retrospective creates or updates L2 then.
 
-ON_DEEP_READ trigger:
-- When AI needs to deeply understand a module but not modify it (e.g., code review, architecture analysis, dependency tracing, explaining complex logic):
-  IF .prizmkit/prizm-docs/<M>/<S>.prizm does not exist:
-    AI reads the source files in S, generates L2 doc for future reference
-- This ensures L2 docs are available for read-heavy analysis tasks, not just modifications
+ON_DEEP_READ behavior:
+- When an AI needs deep understanding but L2 is absent, read the relevant source files directly as fallback context.
+- Create L2 through `/prizmkit-retrospective` after normal development work, or through explicit `/prizmkit-prizm-docs` repair/rebuild when fixing out-of-band doc drift.
+- Do not create L2 merely because a module was read; L2 should capture durable knowledge, not transient investigation notes.

package/bundled/skills/prizmkit-test/SKILL.md

@@ -100,7 +100,7 @@ Interactive mode options:
 
 1. Full project
 2. Single module
-3. Single feature/spec directory
+3. Single historical feature/spec directory
 4. Current change artifact when an artifact directory is available
 
 For `scope=this-change`, build the in-scope source set using this priority:

package/bundled/skills/prizmkit-test/references/boundary-coverage-protocol.md

@@ -4,7 +4,7 @@ Use this protocol during `/prizmkit-test` Phase 4 and update it during Phase 5.
 
 ## Required Boundary Matrix
 
-Build this matrix before writing tests and update it after generated tests pass or produce `needs-review` findings. In `scope=this-change`, build the matrix only for in-scope feature interfaces/endpoints; unrelated interfaces are recorded under `Out of Scope Gaps` and do not trigger generated tests.
+Build this matrix before writing tests and update it after generated tests pass or produce `needs-review` findings. In `scope=this-change`, build the matrix only for interfaces/endpoints in the reviewed change artifact or changed-file set; unrelated interfaces are recorded under `Out of Scope Gaps` and do not trigger generated tests.
 
 | Module | Interface | Service Type | Happy Path | Request Validation | Auth / Permission / Ownership | Domain Invariants | Collection / Pagination | Date / Time | State Transitions | Dependency Failure | Response Contract | N/A Reasons | Final Status |
 |--------|-----------|--------------|------------|--------------------|-------------------------------|-------------------|-------------------------|-------------|-------------------|-------------------|-------------------|-------------|--------------|

package/bundled/skills/prizmkit-test/references/examples.md

@@ -118,7 +118,7 @@ Do not report `user` as covered just because both endpoints have success tests.
 
 **Caller**: `/prizmkit-test scope=this-change artifact_dir=.prizmkit/specs/{{FEATURE_SLUG}}/`
 
-For non-feature changes, use the same command shape with any explicit change artifact directory that contains `spec.md` + `plan.md` and has a changed-files source.
+For non-feature changes, use the same command shape with any explicit change artifact directory that contains `spec.md` + `plan.md` and has a changed-files source. The `{{FEATURE_SLUG}}` path below is a compatibility example for existing feature artifacts; it does not limit `scope=this-change` to feature work.
 
 **Expected behavior**:
 - Reads the change artifact's `spec.md`, `plan.md`, and optional `context-snapshot.md`.
@@ -136,7 +136,7 @@ For non-feature changes, use the same command shape with any explicit change art
 - Generation Policy: in-scope-only
 
 ## Out of Scope Gaps
-- payment: missing legacy tests, but this feature did not change payment code.
+- payment: missing legacy tests, but this change did not affect payment code.
 
 ## Verdict
 PASS

package/bundled/skills/refactor-pipeline-launcher/SKILL.md

@@ -169,7 +169,7 @@ Detect user intent from their message, then follow the corresponding workflow:
    - No (default) — Use defaults for critic review and failure behavior
    - Yes — Configure critic review, stop-on-failure, and reasoning effort options
 
-   Note: Refactor filter defaults to all refactor items (by priority order). If the user selects "Other" on any option, handle their custom input.
+   Note: Refactor filter defaults to all refactor items. Runtime selects eligible refactors in stable list order after dependencies are completed; planners should order unrelated refactors by user priority before launch. If the user selects "Other" on any option, handle their custom input.
 
    **If user chose "Yes" to Advanced config**, run the advanced configuration round (a second `AskUserQuestion` round, plus a reasoning-effort follow-up). It applies to a minority of sessions, so the full question set lives in `${SKILL_DIR}/references/configuration.md` → **Advanced Configuration Round**. Default Critic to Off unless refactor items have `priority: "critical"` (in which case default to On).
 
@@ -225,6 +225,7 @@ Detect user intent from their message, then follow the corresponding workflow:
    - Summarize results: total refactors, succeeded, failed, skipped
    - If all succeeded: each refactor session has already run `prizmkit-retrospective` internally. Ask user what's next.
    - If some failed: show failed refactor IDs and suggest `python3 ./.prizmkit/dev-pipeline/cli.py reset refactor <R-XXX> --clean --run` for a fresh retry
+   - **Browser verification**: Refactor sessions with `browser_interaction` perform supplemental UI verification during execution. After completion, use `python3 ./.prizmkit/dev-pipeline/cli.py refactor status .prizmkit/plans/refactor-list.json` and per-refactor session logs as the source of truth before summarizing browser verification results.
 
    **If background daemon**:
    1. Verify launch:
@@ -328,7 +329,7 @@ Read the error-handling table in `${SKILL_DIR}/references/configuration.md` for
 ### Integration Notes
 
 - **After refactor-planner**: This is the natural next step. When user finishes refactor planning and has `.prizmkit/plans/refactor-list.json`, suggest launching the refactor pipeline.
-- **Session independence**: The pipeline runs completely detached. User can close the AI CLI session, open a new session later, and use this skill to check progress or stop the pipeline.
+- **Session independence**: Only Background daemon mode runs detached. Foreground runs in the current AI CLI session, and Manual mode prints commands without launching.
 - **Single instance per family**: Only one refactor pipeline can run at a time. Different pipeline families may coexist because they use separate daemon metadata and state directories.
 - **Pipeline coexistence**: Refactor pipeline uses `.prizmkit/state/refactor/` separate from `.prizmkit/state/features/` (features) and `.prizmkit/state/bugfix/` (bugs), so all three pipelines can run simultaneously without conflict.
 - **State preservation**: Stopping and restarting the pipeline resumes from where it left off -- completed refactors are not re-run.

package/bundled/skills/refactor-workflow/SKILL.md

@@ -172,10 +172,11 @@ Use Fast Path only after explicit user selection.
 1. Invoke `/prizmkit-plan` with the handoff brief and behavior-preservation expectations.
 2. Invoke `/prizmkit-implement` after the plan is ready.
 3. Run behavior-preservation checks identified in the plan.
-4. Run `/prizmkit-code-review` as the quality and behavior-preservation gate.
-5. Run `/prizmkit-retrospective` before commit because refactors often change durable structure.
-6. Run `/prizmkit-committer` with a `refactor(<scope>):` message.
-7. End the workflow after commit; do not call `refactor-planner`, `refactor-pipeline-launcher`, or monitoring.
+4. Run `/prizmkit-test` only when risk triggers match: public contracts, data models, migration behavior, weak behavior-preservation evidence, or explicit user verification request.
+5. Run `/prizmkit-code-review` when the user requested review, behavior-preservation uncertainty is high, or Full-path-level risk appears during implementation.
+6. Run `/prizmkit-retrospective` only when structure, interfaces, dependencies, behavior, or durable project knowledge changed.
+7. Run `/prizmkit-committer` with a `refactor(<scope>):` message.
+8. End the workflow after commit; do not call `refactor-planner`, `refactor-pipeline-launcher`, or monitoring.
 
 If the refactor starts changing behavior or grows beyond Fast Path criteria, pause and ask whether to switch to `feature-workflow` or Pipeline Path.
 
@@ -241,5 +242,5 @@ For start/stop/logs/retry execution, hand off to `refactor-pipeline-launcher`; d
 
 ## Output
 
-- Fast Path: `spec.md`, `plan.md`, behavior-preserving code changes, behavior-check evidence, review result, Prizm docs sync, and one refactor commit.
+- Fast Path: `spec.md`, `plan.md`, behavior-preserving code changes, behavior-check evidence, conditional test/review evidence when triggered, Prizm docs sync only when durable knowledge changed, and one refactor commit.
 - Pipeline Path: handoff brief, validated `.prizmkit/plans/refactor-list.json` from `refactor-planner`, and pipeline launch/status handled by `refactor-pipeline-launcher`.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "prizmkit",
-  "version": "1.1.112",
+  "version": "1.1.113",
   "description": "Create a new PrizmKit-powered project with clean initialization — no framework dev files, just what you need.",
   "type": "module",
   "bin": {