prizmkit 1.1.196 → 1.1.197

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (106) hide show
  1. package/bin/create-prizmkit.js +0 -2
  2. package/bundled/VERSION.json +3 -3
  3. package/bundled/adapters/antigravity/assets/prizmkit-hook.cjs +1 -1
  4. package/bundled/adapters/antigravity/rules-adapter.js +1 -13
  5. package/bundled/adapters/claude/paths.js +0 -1
  6. package/bundled/adapters/codex/paths.js +0 -1
  7. package/bundled/adapters/codex/settings-adapter.js +1 -1
  8. package/bundled/adapters/cursor/assets/prizmkit-hook.cjs +1 -1
  9. package/bundled/adapters/cursor/rules-adapter.js +1 -13
  10. package/bundled/adapters/pi/paths.js +0 -1
  11. package/bundled/dev-pipeline/.env.example +2 -2
  12. package/bundled/dev-pipeline/README.md +6 -5
  13. package/bundled/dev-pipeline/assets/skill-subagent-integration.md +2 -2
  14. package/bundled/dev-pipeline/prizmkit_runtime/checkpoint_state.py +15 -25
  15. package/bundled/dev-pipeline/prizmkit_runtime/iterate.py +1 -69
  16. package/bundled/dev-pipeline/prizmkit_runtime/platform_detection.py +1 -114
  17. package/bundled/dev-pipeline/prizmkit_runtime/runner_models.py +2 -2
  18. package/bundled/dev-pipeline/prizmkit_runtime/runners.py +78 -0
  19. package/bundled/dev-pipeline/prizmkit_runtime/runtime_helper.py +4 -234
  20. package/bundled/dev-pipeline/prizmkit_runtime/status.py +0 -5
  21. package/bundled/dev-pipeline/prizmkit_runtime/stream_progress.py +9 -1
  22. package/bundled/dev-pipeline/scripts/feature_list_contract.py +0 -43
  23. package/bundled/dev-pipeline/scripts/generate-bootstrap-prompt.py +15 -76
  24. package/bundled/dev-pipeline/scripts/generate-bugfix-prompt.py +9 -112
  25. package/bundled/dev-pipeline/scripts/init-bugfix-pipeline.py +0 -13
  26. package/bundled/dev-pipeline/scripts/init-pipeline.py +0 -1
  27. package/bundled/dev-pipeline/scripts/prompt_framework.py +0 -79
  28. package/bundled/dev-pipeline/scripts/status_common.py +150 -4
  29. package/bundled/dev-pipeline/scripts/update-bug-status.py +20 -2
  30. package/bundled/dev-pipeline/scripts/update-feature-status.py +20 -2
  31. package/bundled/dev-pipeline/scripts/utils.py +6 -0
  32. package/bundled/dev-pipeline/templates/bootstrap-prompt.md +4 -5
  33. package/bundled/dev-pipeline/templates/bootstrap-tier3.md +5 -101
  34. package/bundled/dev-pipeline/templates/bug-fix-list-schema.json +0 -35
  35. package/bundled/dev-pipeline/templates/bugfix-bootstrap-prompt.md +4 -52
  36. package/bundled/dev-pipeline/templates/feature-list-schema.json +0 -25
  37. package/bundled/dev-pipeline/templates/iterate/evaluate-prompt.md +5 -16
  38. package/bundled/dev-pipeline/templates/iterate/evaluation-schema.json +1 -58
  39. package/bundled/dev-pipeline/templates/iterate/headless-plan-prompt.md +1 -12
  40. package/bundled/dev-pipeline/templates/sections/artifact-manifest.md +0 -1
  41. package/bundled/dev-pipeline/templates/sections/bugfix-phase-commit-report.md +0 -1
  42. package/bundled/dev-pipeline/templates/sections/bugfix-phase-diagnose-plan.md +2 -2
  43. package/bundled/dev-pipeline/templates/sections/bugfix-phase-review.md +1 -1
  44. package/bundled/dev-pipeline/templates/sections/bugfix-phase-test.md +1 -1
  45. package/bundled/dev-pipeline/templates/sections/bugfix-reminders.md +0 -2
  46. package/bundled/dev-pipeline/templates/sections/bugfix-session-context.md +0 -1
  47. package/bundled/dev-pipeline/templates/sections/bugfix-task-contract.md +0 -5
  48. package/bundled/dev-pipeline/templates/sections/checkpoint-system.md +1 -2
  49. package/bundled/dev-pipeline/templates/sections/feature-completion-summary.md +1 -1
  50. package/bundled/dev-pipeline/templates/sections/phase-commit-full.md +3 -3
  51. package/bundled/dev-pipeline/templates/sections/phase-prizmkit-test.md +3 -3
  52. package/bundled/dev-pipeline/templates/sections/phase-review-full.md +3 -3
  53. package/bundled/dev-pipeline/templates/sections/phase-specify-plan-full.md +1 -1
  54. package/bundled/dev-pipeline/templates/sections/retrospective.md +3 -3
  55. package/bundled/dev-pipeline/templates/sections/runtime-commit-handoff.md +1 -1
  56. package/bundled/rules/USAGE.md +5 -19
  57. package/bundled/rules/general/coding-baseline.md +2 -2
  58. package/bundled/rules/prizm/prizm-progressive-loading.md +2 -1
  59. package/bundled/skills/_metadata.json +4 -4
  60. package/bundled/skills/app-planner/SKILL.md +1 -1
  61. package/bundled/skills/app-planner/references/generated-plan-review.md +1 -1
  62. package/bundled/skills/app-planner/references/rules-configuration.md +6 -7
  63. package/bundled/skills/app-planner/scripts/validate-rules-configuration.py +2 -20
  64. package/bundled/skills/bug-planner/SKILL.md +5 -6
  65. package/bundled/skills/bug-planner/assets/bug-confirmation-template.md +1 -1
  66. package/bundled/skills/bug-planner/references/input-formats.md +2 -2
  67. package/bundled/skills/bug-planner/references/schema-validation.md +0 -1
  68. package/bundled/skills/bug-planner/scripts/validate-bug-list.py +2 -9
  69. package/bundled/skills/bugfix-pipeline-launcher/SKILL.md +36 -25
  70. package/bundled/skills/bugfix-pipeline-launcher/references/configuration.md +32 -26
  71. package/bundled/skills/feature-pipeline-launcher/SKILL.md +33 -25
  72. package/bundled/skills/feature-pipeline-launcher/references/configuration.md +32 -26
  73. package/bundled/skills/feature-planner/SKILL.md +4 -10
  74. package/bundled/skills/feature-planner/references/completeness-review.md +2 -2
  75. package/bundled/skills/feature-planner/references/new-project-planning.md +0 -1
  76. package/bundled/skills/feature-planner/scripts/validate-and-generate.py +1 -25
  77. package/bundled/skills/iterate-pipeline-launcher/SKILL.md +53 -42
  78. package/bundled/skills/iterate-pipeline-launcher/references/configuration.md +47 -43
  79. package/bundled/skills/prizmkit-code-review/SKILL.md +1 -1
  80. package/bundled/skills/prizmkit-code-review/references/independent-code-review.md +2 -2
  81. package/bundled/skills/prizmkit-implement/SKILL.md +1 -1
  82. package/bundled/skills/prizmkit-init/SKILL.md +6 -7
  83. package/bundled/skills/prizmkit-init/references/config-schema.md +5 -9
  84. package/bundled/skills/prizmkit-init/references/rules/layer-detection.md +1 -1
  85. package/bundled/skills/prizmkit-plan/SKILL.md +1 -1
  86. package/bundled/skills/prizmkit-retrospective/SKILL.md +3 -3
  87. package/bundled/skills/prizmkit-retrospective/references/rule-evolution-steps.md +11 -17
  88. package/bundled/skills/prizmkit-workflow/SKILL.md +6 -3
  89. package/bundled/skills/prizmkit-workflow/references/cold-start-rules.md +8 -7
  90. package/bundled/templates/project-memory-template.md +2 -1
  91. package/package.json +1 -1
  92. package/src/clean.js +29 -7
  93. package/src/config.js +13 -15
  94. package/src/index.js +1 -5
  95. package/src/manifest.js +3 -6
  96. package/src/platforms.js +0 -14
  97. package/src/scaffold.js +93 -313
  98. package/src/upgrade.js +32 -31
  99. package/bundled/adapters/claude/rules-adapter.js +0 -25
  100. package/bundled/adapters/codebuddy/rules-adapter.js +0 -30
  101. package/bundled/adapters/codex/rules-adapter.js +0 -30
  102. package/bundled/adapters/pi/rules-adapter.js +0 -13
  103. package/bundled/dev-pipeline/templates/sections/bugfix-phase-manual-verification.md +0 -10
  104. package/bundled/dev-pipeline/templates/sections/phase-browser-verification.md +0 -111
  105. package/bundled/skills/bug-planner/references/verification.md +0 -26
  106. package/bundled/skills/feature-planner/references/browser-interaction.md +0 -39
@@ -43,10 +43,6 @@ FIXED_RULE_FORBIDDEN_PATTERNS = (
43
43
  )
44
44
 
45
45
  # Output validation constants and contracts
46
- OUTPUT_SOFT_WARN_BYTES = 12 * 1024 # 12 KB
47
- OUTPUT_HARD_MAX_BYTES = 16 * 1024 # 16 KB
48
- MAX_EVOLVED_ENTRIES = 20
49
-
50
46
  UNIFIED_SECTIONS = (
51
47
  "## 1. Confirmed Stack & Scope Decisions",
52
48
  "## 2. Invariant Baseline",
@@ -611,16 +607,6 @@ def _validate_output_file(file_path: Path) -> OutputFileReport:
611
607
  return report
612
608
 
613
609
  report.bytes = len(raw_bytes)
614
- if report.bytes > OUTPUT_HARD_MAX_BYTES:
615
- report.errors.append(
616
- f"raw UTF-8 size {report.bytes} bytes exceeds 16 KB hard limit "
617
- f"({OUTPUT_HARD_MAX_BYTES} bytes)"
618
- )
619
- elif report.bytes >= OUTPUT_SOFT_WARN_BYTES:
620
- report.warnings.append(
621
- f"raw UTF-8 size {report.bytes} bytes reached 12 KB soft warning band "
622
- f"({OUTPUT_SOFT_WARN_BYTES} bytes)"
623
- )
624
610
 
625
611
  try:
626
612
  text = raw_bytes.decode("utf-8")
@@ -732,11 +718,6 @@ def _validate_output_file(file_path: Path) -> OutputFileReport:
732
718
  section5_clean = HTML_COMMENT_RE.sub("", section5)
733
719
  evolved_entries = LIST_ITEM_RE.findall(section5_clean)
734
720
  report.evolved_count = len(evolved_entries)
735
- if report.evolved_count > MAX_EVOLVED_ENTRIES:
736
- report.errors.append(
737
- f"## 5. Project-Evolved Rules has {report.evolved_count} entries "
738
- f"(exceeds hard cap of {MAX_EVOLVED_ENTRIES})"
739
- )
740
721
 
741
722
  _check_contradictions(text, report.errors)
742
723
  return report
@@ -754,7 +735,8 @@ def validate_output(target: Path) -> dict[str, object]:
754
735
  file_reports.append(file_report)
755
736
  elif target.is_dir():
756
737
  entries = sorted(
757
- entry for entry in target.iterdir() if not entry.name.startswith(".")
738
+ entry for entry in target.iterdir()
739
+ if not entry.name.startswith(".") and not (entry.is_dir() and entry.name == "builtin")
758
740
  )
759
741
  if not entries:
760
742
  errors.append(f"no layer-rules files found in directory: {target}")
@@ -37,7 +37,7 @@ Consume only root metadata (`project_description`, `global_context`), `task_coun
37
37
 
38
38
  ## Priority, Severity & Scenario Routing
39
39
 
40
- Preserve exact Severity → Priority mapping (`critical`/`high` → `high`, `medium` → `medium`, `low` → `low`). During bug extraction and prioritization, follow `${SKILL_DIR}/references/severity-rules.md`: do not collapse into uniform `high/high`; require a concrete high-impact indicator (core outage, data loss/integrity, security/auth failure, timeout/crash/500, no workaround) and show `ID | Title | Severity | Severity Rationale | Priority | Priority Rationale | Verification` before writing.
40
+ Preserve exact Severity → Priority mapping (`critical`/`high` → `high`, `medium` → `medium`, `low` → `low`). During bug extraction and prioritization, follow `${SKILL_DIR}/references/severity-rules.md`: do not collapse into uniform `high/high`; require a concrete high-impact indicator (core outage, data loss/integrity, security/auth failure, timeout/crash/500, no workaround) and show `ID | Title | Severity | Severity Rationale | Priority | Priority Rationale` before writing.
41
41
 
42
42
  **Operations**: **Interactive Planning** (`plan bug fixes`, `report bugs`) · **From Log** (`parse errors`, stack traces) · **From Tests** (`these tests are failing`, `parse test output`) · **Validate** (`validate bug list`) · **Summary** (`bug summary`, `list bugs`). Route A generates a new list (`Phase 1-5`); Route B appends via bounded `context` → draft → `append`.
43
43
 
@@ -48,7 +48,7 @@ Gather `project_name`, `project_description`, `global_context` (`tech_stack`, `t
48
48
  **Upfront Material Detection (Hard Rule)**: If materials were supplied upfront, acknowledge them ("I received the following materials: [list]"), read/fetch them immediately, AND still ask "Are there any additional materials you'd like to provide?". Otherwise ask once for code paths, docs/API specs, error screenshots, web links, or log files. Tag bug-specific purposes and normalize cleanly: Material ingestion follows the same durability rule as final output (inline excerpts, omit transient non-code paths, retain purpose-tagged stable refs).
49
49
 
50
50
  ### Phase 2: Bug Collection (CP-BP-2)
51
- Read `${SKILL_DIR}/references/severity-rules.md` and `${SKILL_DIR}/references/input-formats.md` (and `${SKILL_DIR}/references/verification.md` when `verification_type` is `manual`/`hybrid` or involves UI/browser flows) to extract Title, Description (expected vs actual + severity rationale), Severity, Environment, Verification type (`automated` default), Acceptance criteria (verifiable, never subjective), and optional model override.
51
+ Read `${SKILL_DIR}/references/severity-rules.md` and `${SKILL_DIR}/references/input-formats.md` to extract Title, Description (expected vs actual + severity rationale), Severity, Environment, Acceptance criteria (verifiable, never subjective), and optional model override.
52
52
  - **Per-bug clarification**: require confirmed actual/expected behavior, scope, severity rationale, and verification decisions; keep unverified root cause/location as evidence-backed hypotheses or unknowns.
53
53
  - **Per-bug confirmation (mandatory)** — present the structured summary using `${SKILL_DIR}/assets/bug-confirmation-template.md`, ask its three confirmation questions, finalize only after user confirms all three, and ask "Any more bugs to add? (yes/no)".
54
54
 
@@ -87,22 +87,21 @@ Read `${SKILL_DIR}/references/generated-plan-review.md` and follow it as the sin
87
87
 
88
88
  ## Operation: From Log
89
89
  Triggers: `"parse this error log"`, `"here's a stack trace"`, `"parse these errors"`.
90
- 1. Accept log path/content, deduplicate unique errors, and generate entries (Severity via `severity-rules.md`, `error_source`, `verification_type: "automated"`, measurable criteria).
90
+ 1. Accept log path/content, deduplicate unique errors, and generate entries (Severity via `severity-rules.md`, `error_source`, measurable criteria).
91
91
  2. Write `.prizmkit/plans/bug-fix-list.draft.json`. Run the Phase 4 Headless Context Completeness Gate on every parsed bug (clarifying missing expected behavior, scope, rationale, or verification).
92
92
  3. Run merge check (`--dry-run` if list exists) and §Local Generated-Plan Review Gate (`PASS`), insert once, complete §Draft Lifecycle cleanup, and present final result. For later adjustments, recreate the minimum source draft from confirmed in-memory entries or exact-ID `show`, then rerun `generate` or `append --replace-tail` before recommending `bugfix-pipeline-launcher`.
93
93
 
94
94
  ## Operation: From Tests
95
95
  Triggers: `"these tests are failing"`, `"parse test output"`.
96
- 1. Parse failed test cases (`failed_test_path`, `error_message`, `verification_type: "automated"`).
96
+ 1. Parse failed test cases (`failed_test_path`, `error_message`).
97
97
  2. Write `.prizmkit/plans/bug-fix-list.draft.json`. Run the Phase 4 Headless Context Completeness Gate on every parsed failed-test bug (clarifying consequential missing expectations/scope; allow edits/merges/removals).
98
98
  3. Run merge check + §Local Generated-Plan Review Gate (`PASS`), insert once, complete §Draft Lifecycle cleanup, and present final result. For later adjustments, recreate the minimum source draft from confirmed in-memory entries or exact-ID `show`, then rerun `generate` or `append --replace-tail` before recommending `bugfix-pipeline-launcher`.
99
99
 
100
100
  ## Operation: Validate & Operation: Summary
101
101
  - **Validate** (`"validate bug list"`, `"check .prizmkit/plans/bug-fix-list.json"`): run `python3 ${SKILL_DIR}/scripts/validate-bug-list.py validate .prizmkit/plans/bug-fix-list.json` and report errors/warnings without modifying source code.
102
- - **Summary** (`"bug summary"`, `"show bug list"`, `"list bugs"`): print totals by severity/status and prioritized `[B-NNN] Title (SEVERITY) — verification_type` summary.
102
+ - **Summary** (`"bug summary"`, `"show bug list"`, `"list bugs"`): print totals by severity/status and prioritized `[B-NNN] Title (SEVERITY)` summary.
103
103
 
104
104
  ## Verification, Policy, Errors & Output
105
- - **Verification Defaults**: Apply `${SKILL_DIR}/references/verification.md` defaults (`automated`/`manual`/`hybrid`; `browser_interaction` for UI bugs) when not already resolved in Phase 2.
106
105
  - **Next-Step Execution Policy**: Report bug summary/counts, Local generated-plan review result (verdict, reviewed batch IDs plus reviewed root prose when applicable, accepted fixes), single insert `appended_ids` + validation result, Draft cleanup result: `removed`, or retained path and recovery reason, and recommend `bugfix-pipeline-launcher` (or `bug-fix-workflow` for a single interactive bug). Never run scripts directly or auto-execute.
107
106
  - **Error Handling**: Read `${SKILL_DIR}/references/error-recovery.md` on validation failures/interruptions (ask clarifying questions on unparseable logs, ambiguous severity, potential duplicates, empty input, or broken dependencies).
108
107
  - **Output**: `.prizmkit/plans/bug-fix-list.json` conforming to `.prizmkit/dev-pipeline/templates/bug-fix-list-schema.json` (plus validation/summary report when requested).
@@ -8,7 +8,7 @@ Present this after extracting and clarifying each bug:
8
8
  ┌─ Bug Confirmation: B-NNN ─────────────────────────────
9
9
  │ Title: <auto-suggested title>
10
10
  │ Description: <expected vs actual behavior>
11
- │ Severity: <auto-classified> | Verification: <type>
11
+ │ Severity: <auto-classified>
12
12
  │
13
13
  │ Reproduction: <confirmed steps/evidence, or "unknown — bounded discovery required">
14
14
  │ Affected: <evidence-backed module/feature, or "unknown">
@@ -58,7 +58,7 @@ Every new or changed bug is an executable contract for a fresh AI session that h
58
58
 
59
59
  1. **Self-contained contract**: `description` and `acceptance_criteria`, read together, MUST define the objective, current state and actual behavior, desired state and expected behavior, selected scope and decisions, severity rationale, constraints, non-goals, relevant integrations, edge cases, reproduction conditions, verification decisions, and completion evidence. Include the durable severity rationale in `description`. AI rewrite is allowed; dropping a confirmed item is not. There is no word-count or acceptance-criteria count upper limit.
60
60
  2. **Supplementary evidence only**: `user_context` is supplementary evidence only. Logs, stack traces, screenshots, and source excerpts may support diagnosis, but MUST NOT expand, replace, or contradict the contract. If `user_context` were removed, a fresh bug-fix session must still know what fails, where/when it fails, correct behavior, selected scope, and verification. Preserve substantive requirements and constraints in durable prose at matching detail.
61
- 3. **Severity and verification are durable decisions**: Write complete severity rationale, selected scope, environment, reproduction path, verification type, and verification decisions by meaning; never store only an option label or conversational answer.
61
+ 3. **Severity and verification are durable decisions**: Write complete severity rationale, selected scope, environment, reproduction path, and verification expectations by meaning; never store only an option label or conversational answer.
62
62
  4. **Clarify by impact, do not guess**: Clarify expected behavior, selected scope, applicability, and relative references when ambiguity changes the contract. Root cause, code location, and fix approach may stay explicitly labeled as evidence-backed hypotheses or unknowns (`do not force certainty or invent a fix`).
63
63
  5. **Bug isolation**: Keep one bug per task unless bugs share the same root cause or failing contract and the user confirms merging (never domain-merge unrelated bugs in the same module). Repeat shared environment or constraints only when global.
64
64
 
@@ -82,4 +82,4 @@ Every new or changed bug is an executable contract for a fresh AI session that h
82
82
 
83
83
  ## Bounded Existing-List Context (Hard Rule)
84
84
 
85
- For append flows on `.prizmkit/plans/bug-fix-list.json`, NEVER read the complete existing list or NEVER perform default full-history keyword search. Inspect the five most recently appended `{id, description}` pairs from `context` only for material continuation, duplication, conflict, shared public contract, or real prerequisite (`Chronology, broad module overlap, shared technology, reusable implementation ideas, or similar wording alone do not establish relevance`; Add no historical content by default; incorporate only the minimum exact information whose omission would make scope, behavior, compatibility, acceptance criteria, or dependency ordering incorrect; no vague `related to B-123`; add an old ID to `dependencies` only when it is a real execution prerequisite). Use `show --id B-NNN` only for explicit old IDs or validation errors. Write additions-only `{ "bugs": [...] }` (required draft fields: `id`, `title`, `description`, `severity`, `error_source`, `verification_type`, `acceptance_criteria`; `status` defaults to `pending`) in `.prizmkit/plans/bug-fix-list.draft.json`, check `append --dry-run` (Do not ask for separate user approval merely to promote the draft into the final list), pass Local Generated-Plan Review, insert once, delete the draft immediately, and present the final written result. Conditions that must keep a draft live in `references/error-recovery.md`. Preserve unchanged historical bugs.
85
+ For append flows on `.prizmkit/plans/bug-fix-list.json`, NEVER read the complete existing list or NEVER perform default full-history keyword search. Inspect the five most recently appended `{id, description}` pairs from `context` only for material continuation, duplication, conflict, shared public contract, or real prerequisite (`Chronology, broad module overlap, shared technology, reusable implementation ideas, or similar wording alone do not establish relevance`; Add no historical content by default; incorporate only the minimum exact information whose omission would make scope, behavior, compatibility, acceptance criteria, or dependency ordering incorrect; no vague `related to B-123`; add an old ID to `dependencies` only when it is a real execution prerequisite). Use `show --id B-NNN` only for explicit old IDs or validation errors. Write additions-only `{ "bugs": [...] }` (required draft fields: `id`, `title`, `description`, `severity`, `error_source`, `acceptance_criteria`; `status` defaults to `pending`) in `.prizmkit/plans/bug-fix-list.draft.json`, check `append --dry-run` (Do not ask for separate user approval merely to promote the draft into the final list), pass Local Generated-Plan Review, insert once, delete the draft immediately, and present the final written result. Conditions that must keep a draft live in `references/error-recovery.md`. Preserve unchanged historical bugs.
@@ -15,7 +15,6 @@ Use this checklist for manual validation when `validate-bug-list.py` is not avai
15
15
  - [ ] `description`: non-empty string
16
16
  - [ ] `severity`: one of `critical`, `high`, `medium`, `low`
17
17
  - [ ] `error_source.type`: one of `stack_trace`, `user_report`, `failed_test`, `log_pattern`, `monitoring_alert`
18
- - [ ] `verification_type`: one of `automated`, `manual`, `hybrid`
19
18
  - [ ] `acceptance_criteria`: non-empty array of strings
20
19
  - [ ] `status`: one of `pending`, `in_progress`, `completed`, `failed`, `skipped`, `needs_info`, `auto_skipped`; new bugs should start as `pending`, while `auto_skipped` is runtime-managed for blocked descendants
21
20
  - [ ] `dependencies` if present: array of existing `B-NNN` IDs; no self-dependencies
@@ -37,7 +37,6 @@ from datetime import datetime, timezone
37
37
 
38
38
  VALID_SEVERITIES = {"critical", "high", "medium", "low"}
39
39
  VALID_SOURCE_TYPES = {"stack_trace", "user_report", "failed_test", "log_pattern", "monitoring_alert"}
40
- VALID_VERIFICATION_TYPES = {"automated", "manual", "hybrid"}
41
40
  VALID_STATUSES = {"pending", "in_progress", "completed", "failed", "skipped", "needs_info", "auto_skipped"}
42
41
  ALLOWED_LIST_FIELDS = {
43
42
  "$schema", "project_name", "project_description", "created_at",
@@ -50,8 +49,8 @@ ALLOWED_GLOBAL_CONTEXT_FIELDS = {
50
49
  }
51
50
  ALLOWED_BUG_FIELDS = {
52
51
  "id", "title", "description", "severity", "priority", "error_source",
53
- "environment", "affected_modules", "verification_type", "dependencies",
54
- "acceptance_criteria", "status", "user_context", "browser_interaction",
52
+ "environment", "affected_modules", "dependencies",
53
+ "acceptance_criteria", "status", "user_context",
55
54
  "model",
56
55
  }
57
56
  BUG_ID_PATTERN = re.compile(r"^B-\d{3}$")
@@ -236,12 +235,6 @@ def validate(data, feature_ids=None):
236
235
  errors.append("{} ({}): invalid error_source.type '{}' — must be one of {}".format(
237
236
  prefix, bug_id, source_type, sorted(VALID_SOURCE_TYPES)))
238
237
 
239
- # verification_type
240
- vtype = bug.get("verification_type", "")
241
- if vtype not in VALID_VERIFICATION_TYPES:
242
- errors.append("{} ({}): invalid verification_type '{}' — must be one of {}".format(
243
- prefix, bug_id, vtype, sorted(VALID_VERIFICATION_TYPES)))
244
-
245
238
  # acceptance_criteria
246
239
  ac = bug.get("acceptance_criteria", [])
247
240
  if not ac or not isinstance(ac, list):
@@ -1,11 +1,11 @@
1
1
  ---
2
2
  name: "bugfix-pipeline-launcher"
3
- description: "Build bugfix-pipeline commands from a validated .prizmkit/plans/bug-fix-list.json without executing the pipeline. Supports bounded read-only AI CLI model/effort capability probing on demand and produces foreground or background-daemon start commands plus status, logs, stop, and safe retry commands. Use whenever the user wants to start fixing a prepared bug list, construct a bugfix command, check bugfix status, view logs, stop a run, retry a bug, or clean retry B-001. Trigger on: 'start fixing bugs', 'run bugfix pipeline', 'bugfix status', 'bugfix logs', 'stop bug fix', 'retry bug', 'retry B-001'."
3
+ description: "Build bugfix-pipeline commands from a validated .prizmkit/plans/bug-fix-list.json without executing the pipeline. Runs start-time bounded read-only AI CLI model/effort capability probing (skipped on non-start intents) and produces foreground or background-daemon start commands plus status, logs, stop, and safe retry commands. Use whenever the user wants to start fixing a prepared bug list, construct a bugfix command, check bugfix status, view logs, stop a run, retry a bug, or clean retry B-001. Trigger on: 'start fixing bugs', 'run bugfix pipeline', 'bugfix status', 'bugfix logs', 'stop bug fix', 'retry bug', 'retry B-001'."
4
4
  ---
5
5
 
6
6
  # Bugfix Pipeline Command Builder
7
7
 
8
- Build commands for the autonomous bugfix pipeline from within an AI CLI conversation. This skill reads configuration and planning inputs, runs bounded read-only AI CLI capability probes only when custom model or reasoning effort configuration is needed, asks only the choices required to construct an accurate command, displays the command and its effects, then stops.
8
+ Build commands for the autonomous bugfix pipeline from within an AI CLI conversation. On start intents, this skill reads planning inputs, runs bounded start-time read-only AI CLI capability probing after execution-mode selection, displays a Default Parameters Overview Table, asks one focused configuration round (`MODEL`, conditional `PRIZMKIT_EFFORT`, and whether to customize other defaults), displays the command and its effects, then stops. Non-start intents skip CLI probing.
9
9
 
10
10
  ## Execution Boundary
11
11
 
@@ -15,7 +15,7 @@ This skill separates bounded read-only configuration probing from pipeline execu
15
15
  - read a validated bug list and relevant configuration;
16
16
  - build prerequisite-check, start, status, logs, stop, and retry commands;
17
17
  - explain command effects and destructive options before returning them;
18
- - only when the user selects Advanced config or explicitly mentions/asks about a custom model or reasoning effort: run bounded, side-effect-free read-only probes — `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli`, check whether the resolved CLI executable exists on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable). Read-only probes must never invoke prompt runs, interactive logins, file writes, Git mutations, or pipeline state changes.
18
+ - on every Start intent (after Step 2 execution-mode selection): run bounded, side-effect-free read-only probes — `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` (or `AI_CLI=<cli> python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` when a temporary `AI_CLI=<cli>` override is given), route `effort_flag_mode: custom_ignored` directly to Branch 2, otherwise verify `cli_available: true` and executable availability on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable). Read-only probes must never invoke prompt runs, interactive logins, file writes, Git mutations, or pipeline state changes. Non-start intents (`status`, `logs`, `stop`, `retry`) skip read-only probing and record `Read-only CLI probe: skipped (non-start intent)`.
19
19
  - **Prohibited (pipeline execution and state mutation)**:
20
20
  - never execute pipeline commands (`run`, `daemon start`, `stop`, `reset`), start processes, follow logs, change pipeline or Git state, fix source code, or run verification unless explicitly requested by the user.
21
21
 
@@ -27,7 +27,7 @@ Use this skill for:
27
27
 
28
28
  - constructing a foreground or background-daemon bugfix pipeline command;
29
29
  - constructing status, logs, stop, or retry commands;
30
- - translating user-selected runtime options (`MODEL`, adaptive `PRIZMKIT_EFFORT`, retries, logging) into environment variables.
30
+ - translating user-selected runtime options (`MODEL`, adaptive `PRIZMKIT_EFFORT`, and optional default overrides) into environment variables.
31
31
 
32
32
  Do not use it to:
33
33
 
@@ -46,15 +46,17 @@ Accept a user-specified bug-list path when provided. Read the plan directly to s
46
46
 
47
47
  ## Intent Routing
48
48
 
49
+ Classify the request before asking configuration questions:
50
+
49
51
  | Intent | Output |
50
52
  |---|---|
51
- | Start prepared bugs | Prerequisite-check command plus one start command |
52
- | Check status | Daemon status and item-level status commands |
53
- | Show logs | Recent or follow-log command |
54
- | Stop | Daemon stop command plus optional status command |
55
- | Retry failed bug work | State-only, preserve-runtime, fresh-checkout, or clean reset plus a separate run command with safety disclosure |
53
+ | Start prepared bugs | Prerequisite-check command plus one start command (runs start-time read-only CLI probe) |
54
+ | Check status | Daemon status and item-level status commands (`Read-only CLI probe: skipped (non-start intent)`) |
55
+ | Show logs | Recent or follow-log command (`Read-only CLI probe: skipped (non-start intent)`) |
56
+ | Stop | Daemon stop command plus optional status command (`Read-only CLI probe: skipped (non-start intent)`) |
57
+ | Retry failed bug work | State-only, preserve-runtime, fresh-checkout, or clean reset plus a separate run command with safety disclosure (`Read-only CLI probe: skipped (non-start intent)`) |
56
58
 
57
- Only the start intent uses the execution-mode and configuration rounds.
59
+ Only the start intent uses the execution-mode, start-time CLI probe, and configuration rounds. Non-start intents (`status`, `logs`, `stop`, `retry`) skip probing and emit `Read-only CLI probe: skipped (non-start intent)`.
58
60
 
59
61
  ## Start Command Workflow
60
62
 
@@ -62,7 +64,8 @@ Only the start intent uses the execution-mode and configuration rounds.
62
64
 
63
65
  1. Resolve the bug-list path.
64
66
  2. Read the list and summarize total, eligible, blocked, and selected bug IDs.
65
- 3. If an existing run is suspected from user context, include a status command in the output; do not run it.
67
+ 3. Preserve any temporary `AI_CLI=<cli>` override directly; carry it forward into the Default Parameters Overview Table and final command without re-asking.
68
+ 4. If an existing run is suspected from user context, include a status command in the output; do not run it.
66
69
 
67
70
  ### Step 2: Ask execution mode
68
71
 
@@ -71,23 +74,31 @@ Use one `AskUserQuestion` call with exactly one question:
71
74
  - **Foreground (Recommended)** — build `run bugfix`; output remains attached to the user's terminal.
72
75
  - **Background daemon** — build `daemon bugfix start`; the runtime manages detached execution and logs.
73
76
 
74
- Wait for the response before asking configuration.
75
-
76
- ### Step 3: Ask base configuration
77
+ Wait for the response before Step 3.
77
78
 
78
- Use a separate `AskUserQuestion` call with exactly three questions:
79
+ ### Step 3: Probe CLI capabilities, show default parameters, and ask focused configuration
79
80
 
80
- 1. **Verbose logging** — On or Off.
81
- 2. **Max retries** — 3, 1, 5, or a user-provided non-negative integer.
82
- 3. **Advanced config** — No or Yes.
81
+ 1. Load `${SKILL_DIR}/references/configuration.md` and run its start-time bounded read-only AI CLI capability probe sequence first: execute `diagnostics ai-cli` (honoring any temporary `AI_CLI=<cli>` override), consume structured capability fields directly with zero legacy fallback parsing, route `effort_flag_mode: custom_ignored` directly to Branch 2 without letting `<cli> --help` trigger Branch 1, otherwise verify `cli_available: true` and executable availability on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable).
82
+ 2. Display the **Default Parameters Overview Table** before asking configuration questions (noting any pre-specified `AI_CLI` value as already specified):
83
83
 
84
- If Advanced config is No and the user did not specify or ask about a custom model or reasoning effort, skip read-only CLI probing (`diagnostics ai-cli` and `--help`) and record `Read-only CLI probe: skipped (default launch without custom model/effort)`.
84
+ | Parameter | Default | Meaning |
85
+ |---|---|---|
86
+ | `VERBOSE` | `1` | Enable verbose runtime session logging |
87
+ | `MAX_RETRIES` | `3` | Maximum retry attempts per task before marking failed |
88
+ | `STOP_ON_FAILURE` | `1` | Stop subsequent pipeline execution immediately when a task fails (runtime default; override with `STOP_ON_FAILURE=0` to continue to other eligible tasks) |
85
89
 
86
- If Advanced config is Yes — or if the user specified or asked about a custom model or reasoning effort — load `${SKILL_DIR}/references/configuration.md`, run its on-demand read-only CLI capability probe first, and collect confirmed advanced options (stop-on-failure, `MODEL`, and adaptive `PRIZMKIT_EFFORT`) following its four-branch placement rules.
90
+ 3. Then use a separate `AskUserQuestion` call covering only:
91
+ - **`MODEL`** — Default (omit `MODEL` to use the CLI default) or explicit/parameterized model name guided by probe evidence. Under `effort_flag_mode: custom_ignored` or `model_arg_supported: false`, warn clearly per the reference and leave `MODEL` unset. Remember that an individual bug's `"model"` field in `.prizmkit/plans/bug-fix-list.json` takes precedence over global `MODEL`.
92
+ - **`PRIZMKIT_EFFORT`** *(conditional — Branch 4 only)* — ask **only** when Branch 4 of the four ordered adaptive placement branches matches with a non-empty intersection of live `--help` choices and `supported_effort_values`, and no model-embedded intensity is already confirmed. Omit this question and never emit `PRIZMKIT_EFFORT` under Branches 1–3, empty intersection, or model-embedded intensity. If the user supplies a `MODEL` containing embedded intensity in this same Step 3 round under Branch 4, omit `PRIZMKIT_EFFORT` by default and explain that intensity is already encoded in `MODEL` so conflicting settings are not sent together.
93
+ - **Customize other default parameters?** — `No (Recommended — use all defaults from the table)` or `Yes — customize specific defaults`.
94
+ 4. **Conditional customization follow-up (only when Yes)**: if the user selects `Yes — customize specific defaults`, issue one follow-up `AskUserQuestion` call to collect only the specific table parameters to override, validate values (boolean `VERBOSE` / `STOP_ON_FAILURE`, non-negative integer `MAX_RETRIES`), and apply them to the command. Do not ask routine defaults one by one when the user chooses `No`.
87
95
 
88
96
  ### Step 4: Build command fragments
89
97
 
90
- Read `${SKILL_DIR}/references/configuration.md` if not already loaded during Advanced config, and map confirmed choices to environment variables.
98
+ Map confirmed choices to environment variables per `${SKILL_DIR}/references/configuration.md`:
99
+ - When defaults from the table are kept, emit `VERBOSE=1 MAX_RETRIES=3` (foreground) or `--env "VERBOSE=1 MAX_RETRIES=3"` (daemon), and omit runtime-default `STOP_ON_FAILURE=1`.
100
+ - When the user disables stop-on-failure, explicitly emit `STOP_ON_FAILURE=0`. When the user disables verbose logging, emit `VERBOSE=0` or omit `VERBOSE`.
101
+ - Preserve parameterized `MODEL` foreground/daemon quoting rules and any temporary `AI_CLI=<cli>` override.
91
102
 
92
103
  Return this prerequisite-check command separately from the pipeline command:
93
104
 
@@ -95,18 +106,18 @@ Return this prerequisite-check command separately from the pipeline command:
95
106
  python3 ./.prizmkit/dev-pipeline/cli.py diagnostics version && git --version
96
107
  ```
97
108
 
98
- Also include the configured AI CLI check fragment from the reference so the user can verify prerequisites in their own shell. Even when on-demand read-only probing ran during configuration, pipeline commands are never executed by this skill.
109
+ Also include the configured AI CLI check fragment from the reference so the user can verify prerequisites in their own shell. Even though start-time read-only probing ran during configuration, pipeline commands are never executed by this skill.
99
110
 
100
111
  ### Step 5: Return the start command
101
112
 
102
- Foreground template:
113
+ Foreground template (table defaults kept):
103
114
 
104
115
  ```bash
105
116
  VERBOSE=1 MAX_RETRIES=3 \
106
117
  python3 ./.prizmkit/dev-pipeline/cli.py run bugfix .prizmkit/plans/bug-fix-list.json
107
118
  ```
108
119
 
109
- Background template:
120
+ Background template (table defaults kept):
110
121
 
111
122
  ```bash
112
123
  python3 ./.prizmkit/dev-pipeline/cli.py daemon bugfix start .prizmkit/plans/bug-fix-list.json \
@@ -204,7 +215,7 @@ For every request, return:
204
215
  2. Input plan path and selected IDs, when applicable.
205
216
  3. Assumptions and safety warnings.
206
217
  4. Exact command block.
207
- 5. `Read-only CLI probe: <diagnostics-and-help-summary | skipped (default launch without custom model/effort)>`
218
+ 5. `Read-only CLI probe: <diagnostics-and-help-summary | skipped (non-start intent)>`
208
219
  6. `Pipeline execution: not performed by this skill; run any command above manually when ready.`
209
220
 
210
221
  Never report pipeline execution success, process IDs, log locations, bug completion, verification results, or state changes unless the user later supplies authoritative runtime output for explanation.
@@ -1,6 +1,6 @@
1
1
  # Configuration Reference — Bugfix Pipeline Command Builder
2
2
 
3
- Use this reference to run bounded on-demand read-only AI CLI capability probing and translate confirmed user choices into command fragments. This skill constructs and returns pipeline commands; it never executes pipeline runs or mutates state unless explicitly requested by the user.
3
+ Use this reference to run start-time bounded read-only AI CLI capability probing, present the Default Parameters Overview Table, apply the four ordered adaptive placement branches, and translate confirmed user choices into command fragments. This skill constructs and returns pipeline commands; it never executes pipeline runs or mutates state unless explicitly requested by the user.
4
4
 
5
5
  ## Configured AI CLI Prerequisite Command
6
6
 
@@ -14,38 +14,44 @@ python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli
14
14
 
15
15
  `diagnostics ai-cli` reports `detected_cli`, `platform`, `source`, `runtime_mode`, `model_arg_supported`, `effort_flag_mode`, `supported_effort_values`, and `cli_available`, and exits non-zero if no AI CLI can be resolved (`detected_cli: (not found)`).
16
16
 
17
- ## On-Demand Read-Only CLI Capability Probe
17
+ ## Start-Time Read-Only CLI Capability Probe
18
18
 
19
- - **Fast default launch (skip probing)**: When Advanced config is not selected and the user did not mention or ask about a custom model or reasoning effort, skip all read-only CLI probing (`diagnostics ai-cli` and `--help`) and output `Read-only CLI probe: skipped (default launch without custom model/effort)`.
20
- - **On-demand probe trigger**: Run the bounded read-only probe sequence only when Advanced config is selected or the user explicitly specifies or asks about `MODEL` or `PRIZMKIT_EFFORT` (including when paired with a per-invocation `AI_CLI=<cli>` override):
21
- 1. Run `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` — or `AI_CLI=<cli> python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` whenever the user explicitly specifies a per-invocation `AI_CLI=<cli>` override (and preserve `AI_CLI=<cli>` in the returned launch command) — then read the structured capability fields directly: `detected_cli`, `platform`, `source`, `model_arg_supported` (`true`|`false`), `effort_flag_mode` (`flag`|`config`|`none`|`custom_ignored`), `supported_effort_values` (comma-separated values or `none`), and `cli_available` (`true`|`false`). Consume these structured fields directly with zero legacy fallback parsing; if any required capability field is absent, treat the probe as failed under Branch 1.
19
+ - **Non-start intents (skip probing)**: For `status`, `logs`, `stop`, and `retry` intents, skip all read-only CLI probing (`diagnostics ai-cli` and `--help`) and output `Read-only CLI probe: skipped (non-start intent)`.
20
+ - **Start intent probe sequence**: Immediately after Step 2 execution-mode confirmation on every Start intent (before presenting Step 3 model/effort choices), execute this bounded read-only sequence:
21
+ 1. Run `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` — or `AI_CLI=<cli> python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` whenever the user explicitly specifies a temporary `AI_CLI=<cli>` override (and preserve `AI_CLI=<cli>` in the returned launch command) — then read the structured capability fields directly: `detected_cli`, `platform`, `source`, `model_arg_supported` (`true`|`false`), `effort_flag_mode` (`flag`|`config`|`none`|`custom_ignored`), `supported_effort_values` (comma-separated values or `none`), and `cli_available` (`true`|`false`). Consume these structured fields directly with zero legacy fallback parsing; if any required capability field is absent, treat the probe as failed under Branch 1.
22
22
  2. If `effort_flag_mode` is `custom_ignored`, route directly to Branch 2 without requiring `<cli> --help`.
23
23
  3. Otherwise, verify that `cli_available` is `true` and the resolved `detected_cli` executable exists on `PATH` (when not `(not found)`), then run `<cli> --help` (and `<cli> exec --help` when the CLI exposes an `exec` subcommand or `effort_flag_mode` is `config`) to inspect live model flags, independent effort/thinking options, or model-embedded intensity syntax.
24
24
  4. Keep every probe strictly read-only and bounded: never invoke prompt runs, interactive logins, file writes, Git mutations, or pipeline commands (`run`, `daemon start`, `stop`, `reset`). Record a concise summary of observed capability facts in `Read-only CLI probe:`.
25
25
 
26
- ## Base Configuration Mapping
26
+ ## Default Parameters Overview Table
27
27
 
28
- | Choice | Command fragment |
29
- |---|---|
30
- | Verbose Off | `VERBOSE=0` |
31
- | Verbose On | `VERBOSE=1` |
32
- | Max retries N | `MAX_RETRIES=N` |
28
+ Display this table at the start of Step 3 (noting any pre-specified temporary `AI_CLI=<cli>` override as already specified so it is carried forward without re-asking):
33
29
 
34
- ## Advanced Configuration Questions
30
+ | Parameter | Default | Meaning |
31
+ |---|---|---|
32
+ | `VERBOSE` | `1` | Enable verbose runtime session logging (runtime default is `0`; launcher emits `VERBOSE=1` by default) |
33
+ | `MAX_RETRIES` | `3` | Maximum retry attempts per task before marking failed |
34
+ | `STOP_ON_FAILURE` | `1` | Stop subsequent pipeline execution immediately when a task fails (runtime default; omit when kept, emit `STOP_ON_FAILURE=0` to continue to other eligible tasks) |
35
35
 
36
- Ask only when the user selected Advanced config or explicitly requested custom model/effort configuration. Configure `MODEL` before `PRIZMKIT_EFFORT`:
36
+ ## Focused Configuration Flow
37
37
 
38
- 1. **Stop on failure**
39
- - Off (default) — continue to other eligible tasks after a task exhausts retries.
40
- - On — add `STOP_ON_FAILURE=1`.
41
- 2. **Model (`MODEL`)**
38
+ Collect configuration in one focused round after showing the Default Parameters Overview Table. Always configure `MODEL` before `PRIZMKIT_EFFORT`:
39
+
40
+ 1. **Model (`MODEL`)**
42
41
  - Default — omit `MODEL` so the resolved AI CLI uses its default model.
43
42
  - Explicit model — add `MODEL=<name>` (or single-quoted `MODEL='<model-with-level>'` for parameterized model syntax) when `model_arg_supported` is `true` (or when probing failed without diagnostics confirming `model_arg_supported: false` and the user explicitly confirms a model name).
44
43
  - If `diagnostics ai-cli` reports `effort_flag_mode: custom_ignored`, warn clearly that the active custom launch profile ignores both `MODEL` and `PRIZMKIT_EFFORT` (pass provider flags via `.prizmkit/config.json` `ai_cli_launch.extra_args`). If `model_arg_supported: false` is reported for an unresolved/unsupported profile, warn that `MODEL` is not forwarded and leave it unset.
45
44
  - **Per-item precedence**: an individual bug's `"model"` field in `.prizmkit/plans/bug-fix-list.json` takes precedence over the global `MODEL` environment variable during task execution.
46
- 3. **Reasoning effort (`PRIZMKIT_EFFORT`)**
47
- - Default — omit `PRIZMKIT_EFFORT`.
48
- - Adaptive placement — determine whether and how to emit `PRIZMKIT_EFFORT` strictly from the four ordered placement branches below (never hardcode one universal effort list or CLI allowlist).
45
+ 2. **Reasoning effort (`PRIZMKIT_EFFORT`)** *(conditional — Branch 4 only)*
46
+ - Ask and emit `PRIZMKIT_EFFORT` strictly from the four ordered placement branches below (never hardcode one universal effort list or CLI allowlist).
47
+ - Omit the question and never emit `PRIZMKIT_EFFORT` under Branches 1–3, empty capability intersection, or model-embedded intensity syntax.
48
+ - Default under Branch 4 — omit `PRIZMKIT_EFFORT`.
49
+ 3. **Customize other default parameters?**
50
+ - `No (Recommended — use all defaults from the table)` — apply table defaults directly without asking routine parameters one by one.
51
+ - `Yes — customize specific defaults` — ask one follow-up prompt collecting and validating only the specific table overrides:
52
+ - `VERBOSE`: boolean (`1` to enable; `0` or omit `VERBOSE` to disable).
53
+ - `MAX_RETRIES`: non-negative integer `N` (`MAX_RETRIES=N`).
54
+ - `STOP_ON_FAILURE`: boolean (`1` runtime default — omit; `0` explicitly emit `STOP_ON_FAILURE=0` to continue to other eligible tasks after a failure).
49
55
 
50
56
  ## Adaptive Placement Branches (Evaluate 1 → 4 Strictly in Order)
51
57
 
@@ -53,21 +59,21 @@ Evaluate branches strictly from 1 to 4 using `diagnostics ai-cli` capability fie
53
59
 
54
60
  1. **Branch 1 — Probe failed, CLI missing, or insufficient help output**:
55
61
  - Condition: `diagnostics ai-cli` fails or exits non-zero, `detected_cli` is `(not found)` or missing on `PATH`, `cli_available` is `false`, required capability fields (`model_arg_supported`, `effort_flag_mode`, `supported_effort_values`, `cli_available`) are absent, OR (when `effort_flag_mode` is not `custom_ignored`) `<cli> --help` fails or yields insufficient output.
56
- - Action: Never guess or synthesize `PRIZMKIT_EFFORT`. Omit `PRIZMKIT_EFFORT`, state the exact probe gap in `Read-only CLI probe:` and warnings, and allow an optional explicit `MODEL=<name>` only if diagnostics did not confirm `model_arg_supported: false` (otherwise leave `MODEL` unset).
62
+ - Action: Never guess or synthesize `PRIZMKIT_EFFORT`. Omit the `PRIZMKIT_EFFORT` question and variable, state the exact probe gap in `Read-only CLI probe:` and warnings, and allow an optional explicit `MODEL=<name>` only if diagnostics did not confirm `model_arg_supported: false` (otherwise leave `MODEL` unset).
57
63
  2. **Branch 2 — Model-embedded intensity syntax or `custom_ignored` profile**:
58
64
  - Condition: `effort_flag_mode` is `custom_ignored` (routed directly without letting `<cli> --help` failure trigger Branch 1), OR (`model_arg_supported` is `true`, no compatible independent effort mechanism applies — `effort_flag_mode` is `none`, or `supported_effort_values` is `none`, or `effort_flag_mode` is `flag` with no compatible effort/thinking flag in `--help`, excluding compatible `effort_flag_mode: config` — AND live `--help` documents encoding intensity inside the model identifier syntax such as `<model>:<level>` or parameterized model notation).
59
- - Action: Omit `PRIZMKIT_EFFORT`. Under `custom_ignored`, warn clearly that the active custom launch profile ignores both `MODEL` and `PRIZMKIT_EFFORT` (configure provider flags via `.prizmkit/config.json` `ai_cli_launch.extra_args`), never emit `PRIZMKIT_EFFORT` (the runtime rejects non-empty `PRIZMKIT_EFFORT` pre-launch on custom profiles), and omit `MODEL` unless the user explicitly overrides. Under model-embedded intensity syntax, guide the user to set `MODEL=<model-with-level>` (or leave `MODEL` unset for the CLI default).
65
+ - Action: Omit the `PRIZMKIT_EFFORT` question and variable. Under `custom_ignored`, warn clearly that the active custom launch profile ignores both `MODEL` and `PRIZMKIT_EFFORT` (configure provider flags via `.prizmkit/config.json` `ai_cli_launch.extra_args`), never emit `PRIZMKIT_EFFORT` (the runtime rejects non-empty `PRIZMKIT_EFFORT` pre-launch on custom profiles), and leave `MODEL` unset unless the user explicitly overrides. Under model-embedded intensity syntax, guide the user to set `MODEL=<model-with-level>` (or leave `MODEL` unset for the CLI default).
60
66
  3. **Branch 3 — Independent effort explicitly unsupported**:
61
67
  - Condition: Branches 1 and 2 do not match, AND `effort_flag_mode` is `none`, OR `supported_effort_values` is `none`, OR `effort_flag_mode` is `flag` yet live `--help` shows no compatible effort/thinking option (excluding compatible `effort_flag_mode: config`).
62
- - Action: Do not emit `PRIZMKIT_EFFORT` (keep it unset so the runtime does not reject it pre-launch). Still permit `MODEL=<name>` when `model_arg_supported` is `true`.
68
+ - Action: Do not ask for or emit `PRIZMKIT_EFFORT` (keep it unset so the runtime does not reject it pre-launch). Still permit `MODEL=<name>` when `model_arg_supported` is `true`.
63
69
  4. **Branch 4 — Independent effort compatible**:
64
70
  - Condition: (`effort_flag_mode` is `flag` with live `--help` confirming a compatible effort/thinking option, OR `effort_flag_mode` is `config`) AND `supported_effort_values` is non-empty (not `none`).
65
- - Action: When live `--help` enumerates concrete effort/thinking values, intersect those enumerated choices with `supported_effort_values`; if the intersection is empty, omit `PRIZMKIT_EFFORT` and disclose the capability mismatch rather than emitting an unvalidated value. Otherwise present the valid choices and emit `PRIZMKIT_EFFORT=<value>` (or omit for default). If the confirmed `MODEL` value already embeds a reasoning-effort suffix or parameter (for CLIs that accept both model-embedded intensity and an independent effort flag), omit `PRIZMKIT_EFFORT` by default and note that intensity is already specified in `MODEL` so conflicting intensity settings are not sent together.
71
+ - Action: When live `--help` enumerates concrete effort/thinking values, intersect those enumerated choices with `supported_effort_values`; if the intersection is empty, omit the `PRIZMKIT_EFFORT` question and variable and disclose the capability mismatch rather than emitting an unvalidated value. Otherwise present the valid choices and emit `PRIZMKIT_EFFORT=<value>` (or omit for default). If the confirmed `MODEL` value (including one supplied in the same Step 3 round) already embeds a reasoning-effort suffix or parameter, omit `PRIZMKIT_EFFORT` by default and include a conflict-avoidance notice explaining that intensity is already encoded in `MODEL` so conflicting settings are not sent together.
66
72
 
67
73
  ## Command Construction Rules
68
74
 
69
- - Foreground: place environment assignments (`AI_CLI=...` when overridden, `VERBOSE=...`, `MAX_RETRIES=...`, `STOP_ON_FAILURE=...`, `MODEL=...`, `PRIZMKIT_EFFORT=...`) before `python3`.
70
- - Background daemon: place environment assignments inside one `--env "..."` value.
75
+ - Foreground: place environment assignments (`AI_CLI=...` when overridden, `VERBOSE=...`, `MAX_RETRIES=...`, `STOP_ON_FAILURE=0` when disabled, `MODEL=...`, `PRIZMKIT_EFFORT=...`) before `python3`. Default table values emit `VERBOSE=1 MAX_RETRIES=3` and omit `STOP_ON_FAILURE=1` because Runtime defaults `STOP_ON_FAILURE` to `1`.
76
+ - Background daemon: place environment assignments inside one `--env "..."` value. Default table values emit `--env "VERBOSE=1 MAX_RETRIES=3"` and omit `STOP_ON_FAILURE=1`.
71
77
  - **Parameterized `MODEL` quoting**:
72
78
  - Foreground: single-quote parameterized `MODEL` values containing colons, square brackets, parentheses, spaces, commas, or shell metacharacters:
73
79
  ```bash
@@ -1,11 +1,11 @@
1
1
  ---
2
2
  name: "feature-pipeline-launcher"
3
- description: "Build feature-pipeline commands from a validated .prizmkit/plans/feature-list.json without executing the pipeline. Supports bounded read-only AI CLI model/effort capability probing on demand and produces foreground or background-daemon start commands plus status, logs, stop, and safe retry commands. Use whenever the user wants to run the feature pipeline, start building from an existing feature list, construct a feature subset command, inspect pipeline status, view logs, stop a run, or retry a feature. Trigger on: 'run pipeline', 'start pipeline', 'start building', 'build feature list', 'pipeline status', 'pipeline logs', 'stop pipeline', 'retry feature', 'clean retry F-001'."
3
+ description: "Build feature-pipeline commands from a validated .prizmkit/plans/feature-list.json without executing the pipeline. Runs start-time bounded read-only AI CLI model/effort capability probing (skipped on non-start intents) and produces foreground or background-daemon start commands plus status, logs, stop, and safe retry commands. Use whenever the user wants to run the feature pipeline, start building from an existing feature list, construct a feature subset command, inspect pipeline status, view logs, stop a run, or retry a feature. Trigger on: 'run pipeline', 'start pipeline', 'start building', 'build feature list', 'pipeline status', 'pipeline logs', 'stop pipeline', 'retry feature', 'clean retry F-001'."
4
4
  ---
5
5
 
6
6
  # Feature Pipeline Command Builder
7
7
 
8
- Build commands for the autonomous feature pipeline from within an AI CLI conversation. This skill reads configuration and planning inputs, runs bounded read-only AI CLI capability probes only when custom model or reasoning effort configuration is needed, asks only the choices required to construct an accurate command, displays the command and its effects, then stops.
8
+ Build commands for the autonomous feature pipeline from within an AI CLI conversation. On start intents, this skill reads planning inputs, runs bounded start-time read-only AI CLI capability probing after execution-mode selection, displays a Default Parameters Overview Table, asks one focused configuration round (`MODEL`, conditional `PRIZMKIT_EFFORT`, and whether to customize other defaults), displays the command and its effects, then stops. Non-start intents skip CLI probing.
9
9
 
10
10
  ## Execution Boundary
11
11
 
@@ -15,7 +15,7 @@ This skill separates bounded read-only configuration probing from pipeline execu
15
15
  - read a validated feature list and relevant configuration;
16
16
  - build prerequisite-check, start, status, logs, stop, and retry commands;
17
17
  - explain command effects and destructive options before returning them;
18
- - only when the user selects Advanced config or explicitly mentions/asks about a custom model or reasoning effort: run bounded, side-effect-free read-only probes — `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli`, check whether the resolved CLI executable exists on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable). Read-only probes must never invoke prompt runs, interactive logins, file writes, Git mutations, or pipeline state changes.
18
+ - on every Start intent (after Step 2 execution-mode selection): run bounded, side-effect-free read-only probes — `python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` (or `AI_CLI=<cli> python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli` when a temporary `AI_CLI=<cli>` override is given), route `effort_flag_mode: custom_ignored` directly to Branch 2, otherwise verify `cli_available: true` and executable availability on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable). Read-only probes must never invoke prompt runs, interactive logins, file writes, Git mutations, or pipeline state changes. Non-start intents (`status`, `logs`, `stop`, `retry`) skip read-only probing and record `Read-only CLI probe: skipped (non-start intent)`.
19
19
  - **Prohibited (pipeline execution and state mutation)**:
20
20
  - never execute pipeline commands (`run`, `daemon start`, `stop`, `reset`), start processes, follow logs, change pipeline or Git state, implement source code, or run verification unless explicitly requested by the user.
21
21
 
@@ -28,7 +28,7 @@ Use this skill for:
28
28
  - constructing a foreground or background-daemon feature pipeline command;
29
29
  - selecting a feature subset such as `F-001:F-005` or `F-001,F-003`;
30
30
  - constructing status, logs, stop, or retry commands;
31
- - translating user-selected runtime options (`MODEL`, adaptive `PRIZMKIT_EFFORT`, retries, logging) into environment variables.
31
+ - translating user-selected runtime options (`MODEL`, adaptive `PRIZMKIT_EFFORT`, and optional default overrides) into environment variables.
32
32
 
33
33
  Do not use it to:
34
34
 
@@ -51,13 +51,13 @@ Classify the request before asking configuration questions:
51
51
 
52
52
  | Intent | Output |
53
53
  |---|---|
54
- | Start all or selected features | Prerequisite-check command plus one start command |
55
- | Check status | Daemon status and item-level status commands |
56
- | Show logs | Recent or follow-log command |
57
- | Stop | Daemon stop command plus optional status command |
58
- | Retry failed feature work | State-only, preserve-runtime, fresh-checkout, or clean reset plus a separate run command with safety disclosure |
54
+ | Start all or selected features | Prerequisite-check command plus one start command (runs start-time read-only CLI probe) |
55
+ | Check status | Daemon status and item-level status commands (`Read-only CLI probe: skipped (non-start intent)`) |
56
+ | Show logs | Recent or follow-log command (`Read-only CLI probe: skipped (non-start intent)`) |
57
+ | Stop | Daemon stop command plus optional status command (`Read-only CLI probe: skipped (non-start intent)`) |
58
+ | Retry failed feature work | State-only, preserve-runtime, fresh-checkout, or clean reset plus a separate run command with safety disclosure (`Read-only CLI probe: skipped (non-start intent)`) |
59
59
 
60
- Only the start intent uses the execution-mode and configuration rounds.
60
+ Only the start intent uses the execution-mode, start-time CLI probe, and configuration rounds. Non-start intents (`status`, `logs`, `stop`, `retry`) skip probing and emit `Read-only CLI probe: skipped (non-start intent)`.
61
61
 
62
62
  ## Start Command Workflow
63
63
 
@@ -65,7 +65,7 @@ Only the start intent uses the execution-mode and configuration rounds.
65
65
 
66
66
  1. Resolve the feature-list path.
67
67
  2. Read the list and summarize total, eligible, blocked, and selected feature IDs.
68
- 3. Preserve an explicit user feature filter. Do not invent a subset.
68
+ 3. Preserve any pre-specified `--features` subset or temporary `AI_CLI=<cli>` override directly; carry them forward into the Default Parameters Overview Table and final command without re-asking. Do not invent a subset.
69
69
  4. If an existing run is suspected from user context, include a status command in the output; do not run it.
70
70
 
71
71
  ### Step 2: Ask execution mode
@@ -75,23 +75,31 @@ Use one `AskUserQuestion` call with exactly one question:
75
75
  - **Foreground (Recommended)** — build `run feature`; output remains attached to the user's terminal.
76
76
  - **Background daemon** — build `daemon feature start`; the runtime manages detached execution and logs.
77
77
 
78
- Wait for the response before asking configuration.
78
+ Wait for the response before Step 3.
79
79
 
80
- ### Step 3: Ask base configuration
80
+ ### Step 3: Probe CLI capabilities, show default parameters, and ask focused configuration
81
81
 
82
- Use a separate `AskUserQuestion` call with exactly three questions:
82
+ 1. Load `${SKILL_DIR}/references/configuration.md` and run its start-time bounded read-only AI CLI capability probe sequence first: execute `diagnostics ai-cli` (honoring any temporary `AI_CLI=<cli>` override), consume structured capability fields directly with zero legacy fallback parsing, route `effort_flag_mode: custom_ignored` directly to Branch 2 without letting `<cli> --help` trigger Branch 1, otherwise verify `cli_available: true` and executable availability on `PATH`, and inspect `<cli> --help` (plus `<cli> exec --help` when applicable).
83
+ 2. Display the **Default Parameters Overview Table** before asking configuration questions (noting any pre-specified `--features` or `AI_CLI` values as already specified):
83
84
 
84
- 1. **Verbose logging** — On or Off.
85
- 2. **Max retries** — 3, 1, 5, or a user-provided non-negative integer.
86
- 3. **Advanced config** — No or Yes.
85
+ | Parameter | Default | Meaning |
86
+ |---|---|---|
87
+ | `VERBOSE` | `1` | Enable verbose runtime session logging |
88
+ | `MAX_RETRIES` | `3` | Maximum retry attempts per task before marking failed |
89
+ | `STOP_ON_FAILURE` | `1` | Stop subsequent pipeline execution immediately when a task fails (runtime default; override with `STOP_ON_FAILURE=0` to continue to other eligible tasks) |
87
90
 
88
- If Advanced config is No and the user did not specify or ask about a custom model or reasoning effort, skip read-only CLI probing (`diagnostics ai-cli` and `--help`) and record `Read-only CLI probe: skipped (default launch without custom model/effort)`.
89
-
90
- If Advanced config is Yes — or if the user specified or asked about a custom model or reasoning effort — load `${SKILL_DIR}/references/configuration.md`, run its on-demand read-only CLI capability probe first, and collect confirmed advanced options (stop-on-failure, `MODEL`, and adaptive `PRIZMKIT_EFFORT`) following its four-branch placement rules.
91
+ 3. Then use a separate `AskUserQuestion` call covering only:
92
+ - **`MODEL`** — Default (omit `MODEL` to use the CLI default) or explicit/parameterized model name guided by probe evidence. Under `effort_flag_mode: custom_ignored` or `model_arg_supported: false`, warn clearly per the reference and leave `MODEL` unset. Remember that an individual feature's `"model"` field in `.prizmkit/plans/feature-list.json` takes precedence over global `MODEL`.
93
+ - **`PRIZMKIT_EFFORT`** *(conditional — Branch 4 only)* — ask **only** when Branch 4 of the four ordered adaptive placement branches matches with a non-empty intersection of live `--help` choices and `supported_effort_values`, and no model-embedded intensity is already confirmed. Omit this question and never emit `PRIZMKIT_EFFORT` under Branches 1–3, empty intersection, or model-embedded intensity. If the user supplies a `MODEL` containing embedded intensity in this same Step 3 round under Branch 4, omit `PRIZMKIT_EFFORT` by default and explain that intensity is already encoded in `MODEL` so conflicting settings are not sent together.
94
+ - **Customize other default parameters?** — `No (Recommended — use all defaults from the table)` or `Yes — customize specific defaults`.
95
+ 4. **Conditional customization follow-up (only when Yes)**: if the user selects `Yes — customize specific defaults`, issue one follow-up `AskUserQuestion` call to collect only the specific table parameters to override, validate values (boolean `VERBOSE` / `STOP_ON_FAILURE`, non-negative integer `MAX_RETRIES`), and apply them to the command. Do not ask routine defaults one by one when the user chooses `No`.
91
96
 
92
97
  ### Step 4: Build command fragments
93
98
 
94
- Read `${SKILL_DIR}/references/configuration.md` if not already loaded during Advanced config, and map confirmed choices to environment variables.
99
+ Map confirmed choices to environment variables per `${SKILL_DIR}/references/configuration.md`:
100
+ - When defaults from the table are kept, emit `VERBOSE=1 MAX_RETRIES=3` (foreground) or `--env "VERBOSE=1 MAX_RETRIES=3"` (daemon), and omit runtime-default `STOP_ON_FAILURE=1`.
101
+ - When the user disables stop-on-failure, explicitly emit `STOP_ON_FAILURE=0`. When the user disables verbose logging, emit `VERBOSE=0` or omit `VERBOSE`.
102
+ - Preserve parameterized `MODEL` foreground/daemon quoting rules and any temporary `AI_CLI=<cli>` override.
95
103
 
96
104
  Build the prerequisite-check command separately from the pipeline command so the user can inspect each action:
97
105
 
@@ -99,11 +107,11 @@ Build the prerequisite-check command separately from the pipeline command so the
99
107
  python3 ./.prizmkit/dev-pipeline/cli.py diagnostics version && git --version
100
108
  ```
101
109
 
102
- Also include the configured AI CLI check fragment from the reference so the user can verify prerequisites in their own shell. Even when on-demand read-only probing ran during configuration, pipeline commands are never executed by this skill.
110
+ Also include the configured AI CLI check fragment from the reference so the user can verify prerequisites in their own shell. Even though start-time read-only probing ran during configuration, pipeline commands are never executed by this skill.
103
111
 
104
112
  ### Step 5: Return the start command
105
113
 
106
- Foreground template:
114
+ Foreground template (table defaults kept):
107
115
 
108
116
  ```bash
109
117
  VERBOSE=1 MAX_RETRIES=3 \
@@ -117,7 +125,7 @@ VERBOSE=1 MAX_RETRIES=3 \
117
125
  python3 ./.prizmkit/dev-pipeline/cli.py run feature .prizmkit/plans/feature-list.json --features F-001:F-005
118
126
  ```
119
127
 
120
- Background template:
128
+ Background template (table defaults kept):
121
129
 
122
130
  ```bash
123
131
  python3 ./.prizmkit/dev-pipeline/cli.py daemon feature start .prizmkit/plans/feature-list.json \
@@ -222,7 +230,7 @@ For every request, return:
222
230
  2. Input plan path and selected IDs, when applicable.
223
231
  3. Assumptions and safety warnings.
224
232
  4. Exact command block.
225
- 5. `Read-only CLI probe: <diagnostics-and-help-summary | skipped (default launch without custom model/effort)>`
233
+ 5. `Read-only CLI probe: <diagnostics-and-help-summary | skipped (non-start intent)>`
226
234
  6. `Pipeline execution: not performed by this skill; run any command above manually when ready.`
227
235
 
228
236
  Never report pipeline execution success, process IDs, log locations, task completion, verification results, or state changes unless the user later supplies authoritative runtime output for explanation.