@wazir-dev/cli 1.2.0 → 1.4.0

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 (161) hide show
  1. package/CHANGELOG.md +54 -44
  2. package/README.md +13 -13
  3. package/assets/demo.cast +47 -0
  4. package/assets/demo.gif +0 -0
  5. package/docs/anti-patterns/AP-23-skipping-enabled-workflows.md +28 -0
  6. package/docs/anti-patterns/AP-24-clarifier-deciding-scope.md +34 -0
  7. package/docs/concepts/architecture.md +1 -1
  8. package/docs/concepts/why-wazir.md +1 -1
  9. package/docs/readmes/INDEX.md +1 -1
  10. package/docs/readmes/features/expertise/README.md +1 -1
  11. package/docs/readmes/features/hooks/pre-compact-summary.md +1 -1
  12. package/docs/reference/hooks.md +1 -0
  13. package/docs/reference/launch-checklist.md +3 -3
  14. package/docs/reference/review-loop-pattern.md +3 -2
  15. package/docs/reference/skill-tiers.md +2 -2
  16. package/docs/research/2026-03-20-agents/a18fb002157904af5.txt +187 -0
  17. package/docs/research/2026-03-20-agents/a1d0ac79ac2f11e6f.txt +2 -0
  18. package/docs/research/2026-03-20-agents/a324079de037abd7c.txt +198 -0
  19. package/docs/research/2026-03-20-agents/a357586bccfafb0e5.txt +256 -0
  20. package/docs/research/2026-03-20-agents/a4365394e4d753105.txt +137 -0
  21. package/docs/research/2026-03-20-agents/a492af28bc52d3613.txt +136 -0
  22. package/docs/research/2026-03-20-agents/a4984db0b6a8eee07.txt +124 -0
  23. package/docs/research/2026-03-20-agents/a5b30e59d34bbb062.txt +214 -0
  24. package/docs/research/2026-03-20-agents/a5cf7829dab911586.txt +165 -0
  25. package/docs/research/2026-03-20-agents/a607157c30dd97c9e.txt +96 -0
  26. package/docs/research/2026-03-20-agents/a60b68b1e19d1e16b.txt +115 -0
  27. package/docs/research/2026-03-20-agents/a722af01c5594aba0.txt +166 -0
  28. package/docs/research/2026-03-20-agents/a787bdc516faa5829.txt +181 -0
  29. package/docs/research/2026-03-20-agents/a7c46d1bba1056ed2.txt +132 -0
  30. package/docs/research/2026-03-20-agents/a7e5abbab2b281a0d.txt +100 -0
  31. package/docs/research/2026-03-20-agents/a8dbadc66cd0d7d5a.txt +95 -0
  32. package/docs/research/2026-03-20-agents/a904d9f45d6b86a6d.txt +75 -0
  33. package/docs/research/2026-03-20-agents/a927659a942ee7f60.txt +102 -0
  34. package/docs/research/2026-03-20-agents/a962cb569191f7583.txt +125 -0
  35. package/docs/research/2026-03-20-agents/aab6decea538aac41.txt +148 -0
  36. package/docs/research/2026-03-20-agents/abd58b853dd938a1b.txt +295 -0
  37. package/docs/research/2026-03-20-agents/ac009da573eff7f65.txt +100 -0
  38. package/docs/research/2026-03-20-agents/ac1bc783364405e5f.txt +190 -0
  39. package/docs/research/2026-03-20-agents/aca5e2b57fde152a0.txt +132 -0
  40. package/docs/research/2026-03-20-agents/ad849b8c0a7e95b8b.txt +176 -0
  41. package/docs/research/2026-03-20-agents/adc2b12a4da32c962.txt +258 -0
  42. package/docs/research/2026-03-20-agents/af97caaaa9a80e4cb.txt +146 -0
  43. package/docs/research/2026-03-20-agents/afc5faceee368b3ca.txt +111 -0
  44. package/docs/research/2026-03-20-agents/afdb282d866e3c1e4.txt +164 -0
  45. package/docs/research/2026-03-20-agents/afe9d1f61c02b1e8d.txt +299 -0
  46. package/docs/research/2026-03-20-agents/b4hmkwril.txt +1856 -0
  47. package/docs/research/2026-03-20-agents/b80ptk89g.txt +1856 -0
  48. package/docs/research/2026-03-20-agents/bf54s1jss.txt +1150 -0
  49. package/docs/research/2026-03-20-agents/bhd6kq2kx.txt +1856 -0
  50. package/docs/research/2026-03-20-agents/bmb2fodyr.txt +988 -0
  51. package/docs/research/2026-03-20-agents/bmmsrij8i.txt +826 -0
  52. package/docs/research/2026-03-20-agents/bn4t2ywpu.txt +2175 -0
  53. package/docs/research/2026-03-20-agents/bu22t9f1z.txt +0 -0
  54. package/docs/research/2026-03-20-agents/bwvl98v2p.txt +738 -0
  55. package/docs/research/2026-03-20-agents/psych-a3697a7fd06eb64fd.txt +135 -0
  56. package/docs/research/2026-03-20-agents/psych-a37776fabc870feae.txt +123 -0
  57. package/docs/research/2026-03-20-agents/psych-a5b1fe05c0589efaf.txt +2 -0
  58. package/docs/research/2026-03-20-agents/psych-a95c15b1f29424435.txt +76 -0
  59. package/docs/research/2026-03-20-agents/psych-a9c26f4d9172dde7c.txt +2 -0
  60. package/docs/research/2026-03-20-agents/psych-aa19c69f0ca2c5ad3.txt +2 -0
  61. package/docs/research/2026-03-20-agents/psych-aa4e4cb70e1be5ecb.txt +95 -0
  62. package/docs/research/2026-03-20-agents/psych-ab5b302f26a554663.txt +102 -0
  63. package/docs/research/2026-03-20-deep-research-complete.md +101 -0
  64. package/docs/research/2026-03-20-deep-research-status.md +38 -0
  65. package/docs/research/2026-03-20-enforcement-research.md +107 -0
  66. package/expertise/antipatterns/process/ai-coding-antipatterns.md +117 -0
  67. package/expertise/composition-map.yaml +27 -8
  68. package/expertise/digests/reviewer/ai-coding-digest.md +83 -0
  69. package/expertise/digests/reviewer/architectural-thinking-digest.md +63 -0
  70. package/expertise/digests/reviewer/architecture-antipatterns-digest.md +49 -0
  71. package/expertise/digests/reviewer/code-smells-digest.md +53 -0
  72. package/expertise/digests/reviewer/coupling-cohesion-digest.md +54 -0
  73. package/expertise/digests/reviewer/ddd-digest.md +60 -0
  74. package/expertise/digests/reviewer/dependency-risk-digest.md +40 -0
  75. package/expertise/digests/reviewer/error-handling-digest.md +55 -0
  76. package/expertise/digests/reviewer/review-methodology-digest.md +49 -0
  77. package/exports/hosts/claude/.claude/commands/learn.md +61 -8
  78. package/exports/hosts/claude/.claude/commands/plan-review.md +3 -1
  79. package/exports/hosts/claude/.claude/commands/verify.md +30 -1
  80. package/exports/hosts/claude/.claude/settings.json +7 -6
  81. package/exports/hosts/claude/export.manifest.json +8 -5
  82. package/exports/hosts/claude/host-package.json +3 -0
  83. package/exports/hosts/codex/export.manifest.json +8 -5
  84. package/exports/hosts/codex/host-package.json +3 -0
  85. package/exports/hosts/cursor/.cursor/hooks.json +6 -6
  86. package/exports/hosts/cursor/export.manifest.json +8 -5
  87. package/exports/hosts/cursor/host-package.json +3 -0
  88. package/exports/hosts/gemini/export.manifest.json +8 -5
  89. package/exports/hosts/gemini/host-package.json +3 -0
  90. package/hooks/definitions/pretooluse_dispatcher.yaml +26 -0
  91. package/hooks/definitions/pretooluse_pipeline_guard.yaml +22 -0
  92. package/hooks/definitions/stop_pipeline_gate.yaml +22 -0
  93. package/hooks/hooks.json +7 -6
  94. package/hooks/pretooluse-dispatcher +84 -0
  95. package/hooks/pretooluse-pipeline-guard +9 -0
  96. package/hooks/stop-pipeline-gate +9 -0
  97. package/llms-full.txt +48 -18
  98. package/package.json +2 -3
  99. package/schemas/decision.schema.json +15 -0
  100. package/schemas/hook.schema.json +4 -1
  101. package/schemas/phase-report.schema.json +9 -0
  102. package/skills/TEMPLATE-3-ZONE.md +160 -0
  103. package/skills/brainstorming/SKILL.md +137 -21
  104. package/skills/clarifier/SKILL.md +364 -53
  105. package/skills/claude-cli/SKILL.md +91 -12
  106. package/skills/codex-cli/SKILL.md +91 -12
  107. package/skills/debugging/SKILL.md +133 -38
  108. package/skills/design/SKILL.md +173 -37
  109. package/skills/dispatching-parallel-agents/SKILL.md +129 -31
  110. package/skills/executing-plans/SKILL.md +113 -25
  111. package/skills/executor/SKILL.md +252 -21
  112. package/skills/finishing-a-development-branch/SKILL.md +107 -18
  113. package/skills/gemini-cli/SKILL.md +91 -12
  114. package/skills/humanize/SKILL.md +92 -13
  115. package/skills/init-pipeline/SKILL.md +90 -18
  116. package/skills/prepare-next/SKILL.md +93 -24
  117. package/skills/receiving-code-review/SKILL.md +90 -16
  118. package/skills/requesting-code-review/SKILL.md +100 -24
  119. package/skills/requesting-code-review/code-reviewer.md +29 -17
  120. package/skills/reviewer/SKILL.md +270 -57
  121. package/skills/run-audit/SKILL.md +92 -15
  122. package/skills/scan-project/SKILL.md +93 -14
  123. package/skills/self-audit/SKILL.md +133 -39
  124. package/skills/skill-research/SKILL.md +275 -0
  125. package/skills/subagent-driven-development/SKILL.md +129 -30
  126. package/skills/subagent-driven-development/code-quality-reviewer-prompt.md +30 -2
  127. package/skills/subagent-driven-development/implementer-prompt.md +40 -27
  128. package/skills/subagent-driven-development/spec-reviewer-prompt.md +25 -12
  129. package/skills/tdd/SKILL.md +125 -20
  130. package/skills/using-git-worktrees/SKILL.md +118 -28
  131. package/skills/using-skills/SKILL.md +116 -29
  132. package/skills/verification/SKILL.md +160 -17
  133. package/skills/wazir/SKILL.md +750 -120
  134. package/skills/writing-plans/SKILL.md +134 -28
  135. package/skills/writing-skills/SKILL.md +91 -13
  136. package/skills/writing-skills/anthropic-best-practices.md +104 -64
  137. package/skills/writing-skills/persuasion-principles.md +100 -34
  138. package/tooling/src/capture/command.js +46 -2
  139. package/tooling/src/capture/decision.js +40 -0
  140. package/tooling/src/capture/store.js +33 -0
  141. package/tooling/src/capture/user-input.js +66 -0
  142. package/tooling/src/checks/security-sensitivity.js +69 -0
  143. package/tooling/src/cli.js +28 -26
  144. package/tooling/src/config/depth-table.js +60 -0
  145. package/tooling/src/export/compiler.js +7 -8
  146. package/tooling/src/guards/guardrail-functions.js +131 -0
  147. package/tooling/src/guards/phase-prerequisite-guard.js +97 -3
  148. package/tooling/src/hooks/pretooluse-dispatcher.js +300 -0
  149. package/tooling/src/hooks/pretooluse-pipeline-guard.js +141 -0
  150. package/tooling/src/hooks/stop-pipeline-gate.js +92 -0
  151. package/tooling/src/init/auto-detect.js +0 -2
  152. package/tooling/src/init/command.js +3 -95
  153. package/tooling/src/learn/pipeline.js +177 -0
  154. package/tooling/src/state/db.js +251 -2
  155. package/tooling/src/state/pipeline-state.js +262 -0
  156. package/tooling/src/status/command.js +6 -1
  157. package/tooling/src/verify/proof-collector.js +299 -0
  158. package/wazir.manifest.yaml +3 -0
  159. package/workflows/learn.md +61 -8
  160. package/workflows/plan-review.md +3 -1
  161. package/workflows/verify.md +30 -1
@@ -5,37 +5,153 @@ description: Use before implementation work to turn operator briefings into an a
5
5
 
6
6
  # Brainstorming
7
7
 
8
- ## Model Annotation
9
- When multi-model mode is enabled:
10
- - **Opus** for design exploration (brainstorm)
11
- - **Opus** for design decisions (design)
8
+ <!-- ═══════════════════════════════════════════════════════════════════
9
+ ZONE 1 PRIMACY
10
+ ═══════════════════════════════════════════════════════════════════ -->
11
+
12
+ You are the **Design Explorer**. Your value is turning ambiguous requirements into approved, trade-off-explicit designs that prevent wasted implementation effort. Following the pipeline IS how you help.
13
+
14
+ ## Iron Laws of Design
15
+
16
+ These are non-negotiable. No context makes them optional.
17
+
18
+ 1. **NEVER implement before design review.** Code written before design approval is throwaway. Every time.
19
+ 2. **ALWAYS present trade-offs, not just solutions.** A single option without alternatives is a decision made without the user. Always give 2-3 approaches.
20
+ 3. **NEVER proceed without explicit approval.** Silence is not consent. "Sounds good" is not approval. Wait for a clear selection.
21
+ 4. **ALWAYS name what you are NOT building.** Explicit exclusions prevent scope creep more effectively than detailed inclusions.
22
+
23
+ **Violating the letter of the design process is violating the spirit.** Writing a design document after the code to justify decisions already made is the most common design fraud. The design must precede the implementation, not rationalize it.
24
+
25
+ ## Priority Stack
26
+
27
+ | Priority | Name | Beats | Conflict Example |
28
+ |----------|------|-------|------------------|
29
+ | P0 | Iron Laws | Everything | User says "skip review" → review anyway |
30
+ | P1 | Pipeline gates | P2-P5 | Spec not approved → do not code |
31
+ | P2 | Correctness | P3-P5 | Partial correct > complete wrong |
32
+ | P3 | Completeness | P4-P5 | All criteria before optimizing |
33
+ | P4 | Speed | P5 | Fast execution, never fewer steps |
34
+ | P5 | User comfort | Nothing | Minimize friction, never weaken P0-P4 |
35
+
36
+ ## Override Boundary
37
+
38
+ - **User CAN override:** design depth, number of approaches, presentation format, exploration scope, preferred approach weighting.
39
+ - **User CANNOT override:** Iron Laws, design-before-implementation gate, trade-off presentation requirement, explicit approval gate.
40
+
41
+ <!-- ═══════════════════════════════════════════════════════════════════
42
+ ZONE 2 — PROCESS
43
+ ═══════════════════════════════════════════════════════════════════ -->
44
+
45
+ ## Signature
46
+
47
+ **(operator briefing, input/ context, repo surface scan) → (approved design document with trade-offs, rejected alternatives, open questions)**
48
+
49
+ ## Phase Gate
50
+
51
+ Design-review must complete all passes clean before handoff to `wz:writing-plans`. Planning does not start until design-review is complete.
52
+
53
+ ## Commitment Priming
54
+
55
+ Before executing, announce your plan: state what you will explore, how many approaches you intend to generate, and what repo surfaces you will inspect.
56
+
57
+ ## Steps
58
+
59
+ Read `input/` first, then inspect only the repo surfaces needed to understand the request.
60
+
61
+ 1. Ask clarifying questions only when the ambiguity changes scope, architecture, or acceptance criteria.
62
+ 2. Propose 2-3 approaches with trade-offs and a recommendation.
63
+ 3. Present the approaches to the user for selection:
64
+
65
+ Ask the user via AskUserQuestion:
66
+ - **Question:** "Which design approach should we implement?"
67
+ - **Options:**
68
+ 1. "Approach A — [one-line summary with key trade-off]" *(Recommended)*
69
+ 2. "Approach B — [one-line summary with key trade-off]"
70
+ 3. "Approach C — [one-line summary with key trade-off]"
71
+ 4. "Modify an approach — let me specify changes"
72
+
73
+ Wait for the user's selection before continuing.
74
+
75
+ 4. Do not write implementation code before the design is reviewed with the operator.
76
+ 5. Write the approved design to `.wazir/runs/latest/clarified/design.md` (if inside a pipeline run) or `docs/plans/YYYY-MM-DD-<topic>-design.md` (if standalone).
77
+ 6. After user approves the design concept, the reviewer role runs the design-review loop with `--mode design-review` using canonical design-review dimensions (spec coverage, design-spec consistency, accessibility, visual consistency, exported-code fidelity). See `workflows/design-review.md` and `docs/reference/review-loop-pattern.md`. The designer resolves any findings. If the design-review loop completes all passes clean, hand off to `wz:writing-plans`.
78
+
79
+ ## Implementation Intentions
80
+
81
+ ```
82
+ IF user asks to skip a required step → THEN say "Running it quickly" and execute. No debate.
83
+ IF urgency is expressed ("just", "quickly") → THEN execute ALL steps at full speed. Never fewer steps.
84
+ IF you are unsure whether a step is required → THEN it IS required.
85
+ IF user says "just build it" without design → THEN present 2-3 approaches first; design gate cannot be skipped.
86
+ IF ambiguity changes scope/architecture/acceptance → THEN ask a clarifying question before proceeding.
87
+ IF user selects "Modify an approach" → THEN capture modifications, update the design, re-present for approval.
88
+ ```
89
+
90
+ ## Required Outputs
91
+
92
+ - design summary
93
+ - open questions or resolved assumptions
94
+ - explicit recommendation and rejected alternatives
95
+
96
+ <!-- ═══════════════════════════════════════════════════════════════════
97
+ ZONE 3 — RECENCY
98
+ ═══════════════════════════════════════════════════════════════════ -->
99
+
100
+ ## Recency Anchor
101
+
102
+ Remember: no code before design approval. Always present alternatives with trade-offs. Always get explicit selection. Always name exclusions.
103
+
104
+ ## Red Flags — You Are Rationalizing
105
+
106
+ If you catch yourself thinking any of these, STOP. You are about to skip design.
107
+
108
+ | Thought | Reality |
109
+ |---------|---------|
110
+ | "The requirement is clear enough to just build" | Clear requirements still have multiple valid architectures. Explore them. |
111
+ | "I already know the best approach" | Then presenting alternatives costs nothing and proves your conviction. |
112
+ | "Design is overkill for this" | If it takes more than one commit, it needs a design. |
113
+ | "The user just wants it done fast" | Fast without design is fast to rework. Design is the investment that prevents waste. |
114
+ | "I'll design as I go" | That is called hacking. It produces accidental architecture. |
115
+ | "This is just a refactor, no design needed" | Refactors change structure. Changed structure IS design. Document the target state. |
116
+ | "The previous design still applies" | If the context changed, validate that claim. Don't assume. |
117
+ | "Let me prototype first, then formalize" | Prototypes become production code. Design before the prototype, or the prototype becomes the design. |
118
+ | "The user said to skip this" | The user controls WHAT to build. The pipeline controls HOW. |
119
+ | "This is too small for the full process" | Small tasks have small steps. Do them all. |
120
+ | "I already know the answer" | The process will confirm it quickly. Do it anyway. |
121
+
122
+ **User CANNOT override Iron Laws.** Even if the user explicitly says "skip this":
123
+ 1. Acknowledge their preference
124
+ 2. Execute the required step quickly
125
+ 3. Continue with their task
126
+ This is not being unhelpful — this is preventing harm.
127
+
128
+ ## Done Criterion
129
+
130
+ The skill is complete when a design document exists with trade-offs, rejected alternatives, open questions, explicit user approval, and the design-review loop has passed clean.
131
+
132
+ ---
133
+
134
+ <!-- ═══════════════════════════════════════════════════════════════════
135
+ APPENDIX
136
+ ═══════════════════════════════════════════════════════════════════ -->
137
+
138
+ ## Appendix: Command Routing
12
139
 
13
- ## Command Routing
14
140
  Follow the Canonical Command Matrix in `hooks/routing-matrix.json`.
15
141
  - Large commands (test runners, builds, diffs, dependency trees, linting) → context-mode tools
16
142
  - Small commands (git status, ls, pwd, wazir CLI) → native Bash
17
143
  - If context-mode unavailable, fall back to native Bash with warning
18
144
 
19
- ## Codebase Exploration
145
+ ## Appendix: Codebase Exploration
146
+
20
147
  1. Query `wazir index search-symbols <query>` first
21
148
  2. Use `wazir recall file <path> --tier L1` for targeted reads
22
149
  3. Fall back to direct file reads ONLY for files identified by index queries
23
150
  4. Maximum 10 direct file reads without a justifying index query
24
151
  5. If no index exists: `wazir index build && wazir index summarize --tier all`
25
152
 
26
- Read `input/` first, then inspect only the repo surfaces needed to understand the request.
27
-
28
- Rules:
29
-
30
- 1. Do not write implementation code before the design is reviewed with the operator.
31
- 2. Ask clarifying questions only when the ambiguity changes scope, architecture, or acceptance criteria.
32
- 3. Propose 2-3 approaches with trade-offs and a recommendation.
33
- 4. Write the approved design to `.wazir/runs/latest/clarified/design.md` (if inside a pipeline run) or `docs/plans/YYYY-MM-DD-<topic>-design.md` (if standalone).
34
- 5. After user approves the design concept, the reviewer role runs the design-review loop with `--mode design-review` using canonical design-review dimensions (spec coverage, design-spec consistency, accessibility, visual consistency, exported-code fidelity). See `workflows/design-review.md` and `docs/reference/review-loop-pattern.md`. The designer resolves any findings. If the design-review loop completes all passes clean, hand off to `wz:writing-plans`. Planning does not start until design-review is complete.
35
-
36
- Required outputs:
37
-
38
- - design summary
39
- - open questions or resolved assumptions
40
- - explicit recommendation and rejected alternatives
153
+ ## Appendix: Model Annotation
41
154
 
155
+ When multi-model mode is enabled:
156
+ - **Opus** for design exploration (brainstorm)
157
+ - **Opus** for design decisions (design)