@wazir-dev/cli 1.2.0 → 1.4.0

package/llms-full.txt

@@ -1,6 +1,6 @@
 # Wazir — Complete Documentation
 
-> Generated: 2026-03-19T16:28:21Z
+> Generated: 2026-03-20T04:20:52Z
 
 ---
 ## Source: docs/concepts/architecture.md
@@ -17,7 +17,7 @@ Wazir is a host-native engineering OS kit. The host environment (Claude, Codex,
 | Workflows | Phase entrypoints that sequence roles through delivery |
 | Skills | Reusable procedures (wz:tdd, wz:debugging, wz:verification, wz:brainstorming) |
 | Hooks | Guardrails enforcing protected paths, loop caps, and capture routing |
-| Expertise | 268 curated knowledge modules composed into agent prompts |
+| Expertise | 315 curated knowledge modules composed into agent prompts |
 | Templates | Artifact templates for phase outputs and handoff |
 | Schemas | Validation schemas for manifest, hooks, artifacts, and exports |
 | Exports | Generated host packages tailored per supported host |
@@ -93,6 +93,7 @@ open-pencil is integrated as an optional adapter (`open_pencil`) — it is not r
 - [Hooks](../reference/hooks.md)
 - [Expertise & Antipatterns](composition-engine.md)
 
+
 ---
 ## Source: docs/concepts/artifact-model.md
 
@@ -157,6 +158,7 @@ That path must remain gitignored so a normal run still leaves `git status` clean
 - archive stale or disproven learnings instead of rewriting history in place
 - prune external run-state captures when they no longer provide audit or debugging value
 
+
 ---
 ## Source: docs/concepts/composition-engine.md
 
@@ -197,6 +199,7 @@ Brute-force loading of all expertise modules would flood the context window. The
 
 For the complete module listing and anti-pattern catalog, see the [Expertise Index reference](../reference/expertise-index.md).
 
+
 ---
 ## Source: docs/concepts/indexing-and-recall.md
 
@@ -361,6 +364,7 @@ The optional `context-mode` adapter remains:
 
 See the adapter docs for current status and constraints.
 
+
 ---
 ## Source: docs/concepts/observability.md
 
@@ -406,6 +410,7 @@ The capture command family writes under:
 - captured output should reduce context flooding, not increase it
 - summaries must point back to captured files instead of pretending the full output stayed in context
 
+
 ---
 ## Source: docs/concepts/roles-and-workflows.md
 
@@ -471,6 +476,7 @@ If `security_critical: true`, `reviewer` is always included.
 
 Use the files under `roles/` and `workflows/` as the canonical source for role contracts and phase entrypoints. For exact role and workflow tables, see the [Roles Reference](../reference/roles-reference.md).
 
+
 ---
 ## Source: docs/concepts/terminology-policy.md
 
@@ -502,6 +508,7 @@ Do not use terms that describe Wazir as a background service, a web-based contro
 - Use the canonical terms above in all roles, workflows, skills, and documentation.
 - When in doubt, describe what Wazir is, not what it is not.
 
+
 ---
 ## Source: docs/concepts/why-wazir.md
 
@@ -531,7 +538,7 @@ The agent audits its own work in an isolated git worktree. Validates, finds stru
 
 ## 6. Composer
 
-~300 curated expertise modules across 12 domains. The composition engine assembles task-specific agents by loading the right expertise for each role, stack, and concern. The executor building a Flutter RTL app gets Flutter patterns, RTL layout rules, and mobile antipatterns composed into its context. The reviewer gets the corresponding antipattern catalog. Every dispatched agent is a specialist, not a generalist pretending.
+315 curated expertise modules across 12 domains. The composition engine assembles task-specific agents by loading the right expertise for each role, stack, and concern. The executor building a Flutter RTL app gets Flutter patterns, RTL layout rules, and mobile antipatterns composed into its context. The reviewer gets the corresponding antipattern catalog. Every dispatched agent is a specialist, not a generalist pretending.
 
 ## 7. Review Loops
 
@@ -565,6 +572,7 @@ Autonomous phase transition decisions. After each phase, a gating agent reads th
 
 Anti-AI-writing patterns across all text output. A vocabulary blacklist, domain-specific rules, and a self-audit checklist ensure that specs, plans, code comments, commit messages, and documentation read like they were written by a human engineer — not generated by an LLM. Because AI-sounding output erodes trust.
 
+
 ---
 ## Source: docs/getting-started/01-installation.md
 
@@ -647,6 +655,7 @@ npx wazir doctor
 
 [Your First Run](02-first-run.md) — walk through the full pipeline from brief to shipped code.
 
+
 ---
 ## Source: docs/getting-started/02-first-run.md
 
@@ -753,6 +762,7 @@ The 7 steps above map to 14 internal phases:
 - [Roles & Workflows](../concepts/roles-and-workflows.md) — deep dive into role contracts
 - [Composition Engine](../concepts/composition-engine.md) — how expertise modules are loaded
 
+
 ---
 ## Source: docs/guides/memory-and-learnings.md
 
@@ -791,6 +801,7 @@ Wazir keeps learning durable but scoped.
 3. Promote it to `memory/learnings/accepted/` only when the scope and evidence are durable.
 4. Move disproven or obsolete learnings to `memory/learnings/archived/`.
 
+
 ---
 ## Source: docs/guides/troubleshooting.md
 
@@ -896,6 +907,7 @@ If it says the run status is missing:
 - confirm the file exists on disk
 - use `--json` for machine-readable output during automation
 
+
 ---
 ## Source: docs/reference/configuration-reference.md
 
@@ -1100,6 +1112,7 @@ The active manifest currently declares built-in heuristic extractors for:
 - YAML
 - Markdown
 
+
 ---
 ## Source: docs/reference/expertise-index.md
 
@@ -1156,6 +1169,7 @@ The `expertise/humanize/` domain provides AI text pattern detection and removal.
 
 For conceptual understanding of how the composition engine works, see [Composition Engine](../concepts/composition-engine.md).
 
+
 ---
 ## Source: docs/reference/git-flow.md
 
@@ -1203,6 +1217,7 @@ Allowed types: `feat`, `fix`, `docs`, `chore`, `refactor`, `test`, `ci`, `perf`,
 - **CI:** All three validators run on pull requests; `--require-entries` blocks feature/codex/hotfix branches without changelog entries
 - **Roles:** Each role has documented git-flow responsibilities in its contract
 
+
 ---
 ## Source: docs/reference/hooks.md
 
@@ -1294,6 +1309,7 @@ Hook definitions are the authoritative product contracts. The canonical definiti
   - `0` allow
   - `43` block
 
+
 ---
 ## Source: docs/reference/host-exports.md
 
@@ -1348,6 +1364,7 @@ The compiler generates the canonical host packages under `exports/hosts/*`.
 
 The only root host bootstrap retained is `.claude/settings.json`, which mirrors the generated Claude settings contract.
 
+
 ---
 ## Source: docs/reference/launch-checklist.md
 
@@ -1379,7 +1396,7 @@ Submit pull requests to these curated lists (one PR per list, follow each repo's
 ### awesome-claude-code
 - **Repo:** `github.com/anthropics/awesome-claude-code` (or the most-starred community fork)
 - **Section:** Tools / Plugins / Extensions
-- **Entry format:** `[Wazir](https://github.com/MohamedAbdallah-14/Wazir) - Host-native engineering OS kit with 10 roles, 4 phases (15 workflows), and 268 expertise modules.`
+- **Entry format:** `[Wazir](https://github.com/MohamedAbdallah-14/Wazir) - Host-native engineering OS kit with 10 roles, 4 phases (15 workflows), and 315 expertise modules.`
 - **Tips:** Keep the description under 120 characters. Link directly to the repo.
 
 ### awesome-ai-agents
@@ -1409,7 +1426,7 @@ Show HN: Wazir – Engineering OS kit for AI coding agents (Claude, Codex, Gemin
 ### First comment
 Post a comment immediately after submission explaining:
 1. What problem Wazir solves (AI agents lack structured engineering workflows)
-2. How it works (10 canonical roles, 14-phase pipeline, 268 expertise modules)
+2. How it works (10 canonical roles, 4-phase pipeline with 15 workflows, 315 expertise modules)
 3. What makes it different (host-native, works across Claude/Codex/Gemini/Cursor)
 4. Quick install: `npx @wazir-dev/cli init`
 5. Invite feedback -- HN readers appreciate genuine requests for input
@@ -1453,7 +1470,7 @@ Structure as a 5-7 tweet thread:
 
 1. **Hook tweet:** One-liner about the problem + link to repo.
 2. **What it is:** Brief description of Wazir.
-3. **Architecture:** 10 roles, 4 phases (15 workflows), 308 modules (include a diagram image).
+3. **Architecture:** 10 roles, 4 phases (15 workflows), 315 modules (include a diagram image).
 4. **Demo:** Short GIF or screenshot of a workflow in action.
 5. **Multi-host:** Works with Claude, Codex, Gemini, and Cursor.
 6. **Install:** `npx @wazir-dev/cli init`
@@ -1524,6 +1541,7 @@ Monitor these metrics weekly for the first month, then monthly:
 | External PRs | 2+ |
 | HN points | 50+ |
 
+
 ---
 ## Source: docs/reference/marketplace-listings.md
 
@@ -1604,6 +1622,7 @@ Run through this checklist after every `npm publish`:
 - [ ] **Host exports:** Run `npx wazir export --check` to verify no drift
 - [ ] **CHANGELOG:** Verify `CHANGELOG.md` is updated with the new version entry
 
+
 ---
 ## Source: docs/reference/release-process.md
 
@@ -1642,6 +1661,7 @@ When no Wazir release tag exists yet:
 - Legacy tags are not considered release boundaries
 - The first release tag will be `v1.0.0` (or `v0.1.0` if pre-stable)
 
+
 ---
 ## Source: docs/reference/review-loop-pattern.md
 
@@ -2184,6 +2204,7 @@ Every phase exit produces a report saved to `.wazir/runs/latest/reviews/<phase>-
 6. **Context Savings** — context-mode stats if available, omit section if not
 7. **Time Spent** — wall-clock elapsed time from phase start to end — log "codex marker not found in output, cannot extract findings" and present a warning to the user with 0 findings extracted. The raw file is preserved for manual review. Do NOT fall back to `tail` or any best-effort extraction that could leak traces into context.
 
+
 ---
 ## Source: docs/reference/roles-reference.md
 
@@ -2266,6 +2287,7 @@ Roles that explore broadly (clarifier, researcher, planner) benefit most from L1
 
 See [Indexing and Recall](../concepts/indexing-and-recall.md) for full details on tiers and commands.
 
+
 ---
 ## Source: docs/reference/skill-tiers.md
 
@@ -2417,6 +2439,7 @@ If superpowers or Claude Code introduces a composition/layering mechanism in the
 
 5. **The `wz:` prefix is already applied** in skill names within the Wazir SKILL.md frontmatter for all forked skills, consistent with the Own tier convention.
 
+
 ---
 ## Source: docs/reference/skills.md
 
@@ -2454,6 +2477,7 @@ These skills remain on the active surface:
 - Skills must not instruct users to run background services or wrapper scripts that are not part of the canonical workflow surface.
 - When a skill becomes contradictory to the current operating model, remove it from `skills/`.
 
+
 ---
 ## Source: docs/reference/templates.md
 
@@ -2487,6 +2511,7 @@ Each template requires run metadata, sources, loop number, and approval status w
 
 Schema-backed examples under `templates/examples/` exist to keep schemas, examples, and validation in sync.
 
+
 ---
 ## Source: docs/reference/tooling-cli.md
 
@@ -2589,6 +2614,7 @@ Executable documentation claims are registered in:
 
 `wazir validate docs` uses that file plus active markdown link checks to prevent stale command and path claims from silently drifting.
 
+
 ---
 ## Source: README.md
 
@@ -2626,7 +2652,7 @@ Executable documentation claims are registered in:
 I'm Mohamed Abdallah. I kept watching AI agents write confident code that broke in production, skip tests, and forget what we agreed on yesterday. So I stopped asking them to be better and built them an engineering department instead.
 
 **Wazir puts engineering discipline inside AI coding agents.**
-No wrapper. No server. Just structure -- inside Claude, Codex, Gemini, and Cursor. Built on 300+ research sources distilled into 268 curated expertise modules across 12 domains.
+No wrapper. No server. Just structure -- inside Claude, Codex, Gemini, and Cursor. Built on 300+ research sources distilled into 315 curated expertise modules across 12 domains.
 
 ---
 
@@ -2671,7 +2697,7 @@ AI agents love to announce they're finished. Wazir doesn't care. Every phase loo
 
 ## The Pipeline
 
-Every task flows through 14 phases. Three are adversarial review gates that block progress until the reviewer explicitly approves. Rejection loops back to the authoring phase.
+Every task flows through 15 workflows grouped into 4 phases. Three are adversarial review gates that block progress until the reviewer explicitly approves. Rejection loops back to the authoring phase.
 
 ```mermaid
 graph LR
@@ -2718,7 +2744,7 @@ Three concepts.
 
 **2 -- Phases are artifact checkpoints, not conversation stages.** Every phase consumes a named artifact from the previous phase and produces a named artifact for the next. Nothing flows through conversation history. A session can end, a new agent can pick up the artifacts, and delivery continues. The handoff is explicit, structured, and schema-validated against 19 JSON schemas. See [Architecture](docs/concepts/architecture.md).
 
-**3 -- The composition engine loads the right expert automatically.** One agent pretending to be an expert in everything is an expert in nothing. A 4-layer system (always, auto, stacks, concerns) decides which of 268 expertise modules load into each role's context. The executor gets modules on how to build. The verifier gets modules on what to detect. The reviewer gets modules on what to flag. All resolved automatically from the task's declared stack and concerns. Max 15 modules per dispatch, token budget enforced.
+**3 -- The composition engine loads the right expert automatically.** One agent pretending to be an expert in everything is an expert in nothing. A 4-layer system (always, auto, stacks, concerns) decides which of 315 expertise modules load into each role's context. The executor gets modules on how to build. The verifier gets modules on what to detect. The reviewer gets modules on what to flag. All resolved automatically from the task's declared stack and concerns. Max 15 modules per dispatch, token budget enforced.
 
 ---
 
@@ -2763,15 +2789,15 @@ Run `wazir capture usage` at the end of a session to see the savings:
 
 **10 canonical role contracts.** Clarifier, researcher, specifier, content-author, designer, planner, executor, verifier, reviewer, learner. Each has enforceable inputs, outputs, and escalation rules. [Roles reference](docs/reference/roles-reference.md)
 
-**Adversarial review at three chokepoints.** Spec-challenge, plan-review, and final review run by the reviewer role, never the phase author. Nine hard approval gates span the 14-phase pipeline. Nothing advances without explicit clearance. [Architecture](docs/concepts/architecture.md)
+**Adversarial review at three chokepoints.** Spec-challenge, plan-review, and final review run by the reviewer role, never the phase author. Nine hard approval gates span the 15-workflow pipeline. Nothing advances without explicit clearance. [Architecture](docs/concepts/architecture.md)
 
-**268 curated expertise modules across 12 domains.** Loaded selectively per role per phase via a 4-layer composition engine. Max 15 modules per dispatch, token budget enforced. Wazir ships with 268. Yours could be next. [Expertise index](docs/reference/expertise-index.md)
+**315 curated expertise modules across 12 domains.** Loaded selectively per role per phase via a 4-layer composition engine. Max 15 modules per dispatch, token budget enforced. Wazir ships with 315. Yours could be next. [Expertise index](docs/reference/expertise-index.md)
 
 **Three-tier recall for token savings.** L0 (~~100 tokens), L1 (~~500-2k tokens), direct read for full source. Symbol-first exploration searches the index before reading source. Capture routing redirects large tool output to files. Result: 60-80% token reduction on exploration-heavy phases, measured per-session by `wazir capture usage`. [Indexing and Recall](docs/concepts/indexing-and-recall.md)
 
 **Structured learning.** Proposed learnings require explicit review and scope tagging before promotion. Only learnings whose file patterns overlap the current task get injected into context. The system improves per-project without drifting.
 
-**7 hook contracts for structural guardrails.** These enforce protected path writes (exit 42), loop caps (exit 43), and session observability. [Hooks](docs/reference/hooks.md)
+**8 hook contracts for structural guardrails.** These enforce protected path writes (exit 42), loop caps (exit 43), and session observability. [Hooks](docs/reference/hooks.md)
 
 **20+ callable skills.** `/wazir` runs the full pipeline. `/wazir audit security` runs a codebase audit. `/wazir prd` generates a product requirements document from completed runs. Plus TDD, verification, debugging, and more -- each enforcing an exact procedure with evidence at every step. [Skills](docs/reference/skills.md)
 
@@ -2783,19 +2809,19 @@ Run `wazir capture usage` at the end of a session to see the savings:
 
 ## Compared to Other Tools
 
-The AI coding tool space is fragmenting. Developers bolt together separate plugins for workflow management, specification, memory, output compression, and orchestration. Not every project needs 14 phases. For a weekend hack, prompting is fine. For production, you want structure.
+The AI coding tool space is fragmenting. Developers bolt together separate plugins for workflow management, specification, memory, output compression, and orchestration. Not every project needs 15 workflows. For a weekend hack, prompting is fine. For production, you want structure.
 
 
 | Dimension              | Wazir                         | [Superpowers](https://github.com/obra/superpowers) | [Spec-Kit](https://github.com/github/spec-kit) | [Micro-Agent](https://github.com/BuilderIO/micro-agent) | [Distill](https://github.com/samuelfaj/distill) | [Claude-Mem](https://github.com/thedotmack/claude-mem) | [OMC](https://github.com/yeachan-heo/oh-my-claudecode) |
 | ---------------------- | ----------------------------- | -------------------------------------------------- | ---------------------------------------------- | ------------------------------------------------------- | ----------------------------------------------- | ------------------------------------------------------ | ------------------------------------------------------ |
 | **Category**           | Engineering OS                | Skills framework                                   | Spec toolkit                                   | Code gen agent                                          | Output compressor                               | Memory plugin                                          | Orchestration layer                                    |
-| **Scope**              | Full lifecycle (14 phases)    | Dev workflow (~20 skills)                          | Specify / Plan / Implement                     | Single-file TDD loop                                    | CLI output compression                          | Session memory                                         | Multi-agent orchestration                              |
+| **Scope**              | Full lifecycle (15 workflows) | Dev workflow (~20 skills)                          | Specify / Plan / Implement                     | Single-file TDD loop                                    | CLI output compression                          | Session memory                                         | Multi-agent orchestration                              |
 | **Enforced roles**     | 10 canonical, contractual     | None (skills only)                                 | None                                           | None                                                    | None                                            | None                                                   | 32 agents (behavioral)                                 |
-| **Phase model**        | 14 explicit, artifact-gated   | 7-step (advisory)                                  | 3-step                                         | 1 (generate/test)                                       | N/A                                             | N/A                                                    | 5-step pipeline                                        |
+| **Phase model**        | 15 explicit, artifact-gated   | 7-step (advisory)                                  | 3-step                                         | 1 (generate/test)                                       | N/A                                             | N/A                                                    | 5-step pipeline                                        |
 | **Adversarial review** | 3 gate phases                 | Code review skill                                  | No                                             | No                                                      | No                                              | No                                                     | team-verify step                                       |
 | **Context management** | L0/L1 tiered recall           | None                                               | None                                           | None                                                    | LLM compression                                 | Vector DB (ChromaDB)                                   | Token routing                                          |
 | **Schema validation**  | 19 JSON schemas               | No                                                 | No                                             | No                                                      | No                                              | No                                                     | No                                                     |
-| **Guardrails**         | 7 hook contracts              | None                                               | None                                           | None                                                    | None                                            | 5 hooks (memory)                                       | Agent tracking                                         |
+| **Guardrails**         | 8 hook contracts              | None                                               | None                                           | None                                                    | None                                            | 5 hooks (memory)                                       | Agent tracking                                         |
 | **External deps**      | None (host-native)            | None (prompt-only)                                 | Python CLI                                     | Node.js CLI                                             | Node.js + LLM                                   | ChromaDB, SQLite, Bun                                  | tmux, exp. teams API                                   |
 | **Host support**       | Claude, Codex, Gemini, Cursor | Claude, Codex, Gemini, Cursor, OpenCode            | Claude, Copilot, Gemini                        | Any LLM provider                                        | Any LLM                                         | Claude Code only                                       | Claude Code (+ workers)                                |
 
@@ -2858,8 +2884,8 @@ The pipeline, roles, and expertise modules are stable and used in production by
 
 What's solid:
 
-- The 14-phase pipeline and 10 role contracts
-- 268 expertise modules across 12 domains
+- The 4-phase pipeline (15 workflows) and 10 role contracts
+- 315 expertise modules across 12 domains
 - Host exports for Claude, Codex, Gemini, and Cursor
 - The composition engine and tiered recall system
 
@@ -2906,6 +2932,7 @@ See [CONTRIBUTING.md](CONTRIBUTING.md) for development setup, branch conventions
 ## License
 
 MIT -- see [LICENSE](LICENSE).
+
 ---
 ## Source: CONTRIBUTING.md
 
@@ -3011,6 +3038,7 @@ Not sure where to start? Open a [Discussion](https://github.com/MohamedAbdallah-
 4. **Merge:** Once approved and all checks pass, a maintainer will merge your PR using a squash merge with a conventional commit message.
 5. **Timeline:** We aim to provide initial review feedback within a few days. If your PR has been open for more than a week without a response, feel free to leave a comment or ping in Discussions.
 
+
 ---
 ## Source: AGENTS.md
 
@@ -3126,3 +3154,5 @@ This project uses Codex as a secondary reviewer. Review artifacts are in `tasks/
 - Use isolated feature branches
 - Reference `wazir.manifest.yaml` for the project manifest and schema
 
+
+---

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@wazir-dev/cli",
-  "version": "1.2.0",
+  "version": "1.4.0",
   "type": "module",
   "description": "Host-native engineering OS kit for AI coding agents — roles, phases, expertise modules, quality gates for Claude, Codex, Gemini & Cursor",
   "bin": {
@@ -29,7 +29,7 @@
   ],
   "scripts": {
     "test": "npm run test:active",
-    "test:active": "node --test tooling/test/cli.test.js tooling/test/validate.test.js tooling/test/index.test.js tooling/test/doctor-status.test.js tooling/test/guard-hooks.test.js tooling/test/export.test.js tooling/test/capture.test.js tooling/test/schema-examples.test.js tooling/test/git-flow-docs.test.js tooling/test/role-contracts.test.js tooling/test/ci-workflow.test.js tooling/test/contributing.test.js tooling/test/author-artifact.test.js tooling/test/capture/usage.test.js tooling/test/adapters/composition-engine.test.js tooling/test/docs-drift.test.js tooling/test/init.test.js tooling/test/gating/agent.test.js tooling/test/guards/phase-prerequisite-guard.test.js tooling/test/hooks/context-mode-router.test.js tooling/test/hooks/session-start.test.js tooling/test/input-scanning.test.js tooling/test/init/auto-detect.test.js tooling/test/adapters/model-router.test.js tooling/test/state/db.test.js tooling/test/reports/phase-report.test.js tooling/test/logo-svg.test.js tooling/test/adapters/context-mode.test.js",
+    "test:active": "node --test tooling/test/cli.test.js tooling/test/validate.test.js tooling/test/index.test.js tooling/test/doctor-status.test.js tooling/test/guard-hooks.test.js tooling/test/export.test.js tooling/test/capture.test.js tooling/test/schema-examples.test.js tooling/test/git-flow-docs.test.js tooling/test/role-contracts.test.js tooling/test/ci-workflow.test.js tooling/test/contributing.test.js tooling/test/author-artifact.test.js tooling/test/capture/usage.test.js tooling/test/adapters/composition-engine.test.js tooling/test/docs-drift.test.js tooling/test/init.test.js tooling/test/gating/agent.test.js tooling/test/guards/phase-prerequisite-guard.test.js tooling/test/hooks/context-mode-router.test.js tooling/test/hooks/session-start.test.js tooling/test/input-scanning.test.js tooling/test/init/auto-detect.test.js tooling/test/adapters/model-router.test.js tooling/test/state/db.test.js tooling/test/reports/phase-report.test.js tooling/test/logo-svg.test.js tooling/test/adapters/context-mode.test.js tooling/test/verify/proof-collector.test.js tooling/test/state/pipeline-state.test.js tooling/test/guards/guardrail-functions.test.js tooling/test/hooks/stop-pipeline-gate.test.js tooling/test/hooks/pretooluse-pipeline-guard.test.js tooling/test/hooks/pretooluse-dispatcher.test.js tooling/test/config/depth-table.test.js tooling/test/state/artifact-dependencies.test.js tooling/test/integration/three-layer-enforcement.test.js tooling/test/learn/pipeline.test.js tooling/test/reports/command.test.js tooling/test/state/command.test.js",
     "test:coverage": "c8 npm run test:active",
     "prepare": "husky"
   },
@@ -61,7 +61,6 @@
     "tdd"
   ],
   "dependencies": {
-    "@inquirer/prompts": "^8.3.2",
     "ajv": "^8.18.0",
     "yaml": "^2.0.0"
   },

package/schemas/decision.schema.json

@@ -0,0 +1,15 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://wazir.dev/schemas/decision.schema.json",
+  "title": "Decision Log Entry",
+  "type": "object",
+  "required": ["timestamp", "phase", "decision", "reason"],
+  "properties": {
+    "timestamp": { "type": "string", "format": "date-time" },
+    "phase": { "type": "string", "minLength": 1 },
+    "decision": { "type": "string", "maxLength": 200 },
+    "reason": { "type": "string", "maxLength": 500 },
+    "task_id": { "type": "string" }
+  },
+  "additionalProperties": false
+}

package/schemas/hook.schema.json

@@ -24,7 +24,10 @@
         "stop_handoff_harvest",
         "protected_path_write_guard",
         "loop_cap_guard",
-        "context_mode_router"
+        "context_mode_router",
+        "stop_pipeline_gate",
+        "pretooluse_pipeline_guard",
+        "pretooluse_dispatcher"
       ]
     },
     "description": { "type": "string", "minLength": 1 },

package/schemas/phase-report.schema.json

@@ -6,6 +6,15 @@
   "required": ["phase_name", "run_id", "timestamp", "attempted_actions", "drift_analysis", "quality_metrics", "risk_flags", "decisions"],
   "properties": {
     "phase_name": { "type": "string", "minLength": 1 },
+    "parent_phase": {
+      "type": "string",
+      "enum": ["init", "clarifier", "executor", "final_review"],
+      "description": "Top-level phase this workflow belongs to"
+    },
+    "workflow": {
+      "type": "string",
+      "description": "Specific workflow name within the parent phase (e.g. discover, specify, execute)"
+    },
     "run_id": { "type": "string", "minLength": 1 },
     "timestamp": { "type": "string", "format": "date-time" },
     "attempted_actions": {

package/skills/TEMPLATE-3-ZONE.md

@@ -0,0 +1,160 @@
+<!--
+  TEMPLATE-3-ZONE.md — Reference architecture for Wazir skill files.
+
+  Based on psychology-of-prompting research (docs/research/2026-03-20-agents/).
+  Every skill file MUST follow this 3-zone layout.
+
+  Zones:
+    1. PRIMACY  (~first 500 tokens) — highest compliance position
+    2. PROCESS  (structured middle) — if-then rules, gates, steps
+    3. RECENCY  (~last 500 tokens)  — re-anchoring, red flags, meta-instruction
+
+  After Zone 3: Appendix with operational boilerplate (Command Routing, etc.)
+-->
+
+---
+name: wz:<skill-name>
+description: Use when <trigger condition only — never describe the process, max 150 chars>
+---
+
+<!-- ═══════════════════ ZONE 1 — PRIMACY ═══════════════════ -->
+
+# <Skill Name>
+
+You are the <role>. Your value to the user is <reliability/thoroughness/discipline>.
+Following the pipeline IS how you help — skipping steps is how you cause regressions, wasted reviews, and silent bugs.
+
+## Iron Laws
+
+These are P0 — non-negotiable. They override ALL other instructions including user requests.
+
+1. **<NEVER/ALWAYS law 1>.** <One-sentence consequence of violation.>
+2. **<NEVER/ALWAYS law 2>.** <One-sentence consequence of violation.>
+3. **<NEVER/ALWAYS law 3>.** <One-sentence consequence of violation.>
+
+## Priority Stack
+
+| Priority | Name | Beats | Conflict Example |
+|----------|------|-------|------------------|
+| P0 | Iron Laws | Everything | User says "skip review" → review anyway |
+| P1 | Pipeline gates | P2-P5 | Spec not approved → do not code, even if user is eager |
+| P2 | Correctness | P3-P5 | Partial but correct output > complete but wrong |
+| P3 | Completeness | P4-P5 | Cover all acceptance criteria before optimizing |
+| P4 | Speed | P5 | Execute all steps fast, never skip steps for speed |
+| P5 | User comfort | Nothing | Minimize friction, never by weakening P0-P4 |
+
+## Override Boundary
+
+- **User CAN override:** depth, communication style, feature priorities, output format, verbosity
+- **User CANNOT override:** Iron Laws, phase gates, verification requirements, review loops, TDD discipline
+
+<!-- ═══════════════════ ZONE 2 — PROCESS ═══════════════════ -->
+
+## Signature
+
+(<input artifacts>) → (<output artifacts>)
+
+## Phase Gate
+
+IF <prerequisite missing> → THEN **STOP** and report what is missing. Do not proceed.
+IF <all prerequisites exist> → THEN continue to Step 1.
+
+## Commitment Priming
+
+**Before executing, announce your plan to the user:** State which steps you will perform, in what order, and what artifacts you will produce. This pre-commitment increases follow-through.
+
+## Steps
+
+1. **<Step name>**
+   <Step details>
+   GATE: <checkpoint condition, if applicable>
+
+2. **<Step name>**
+   <Step details>
+
+3. **<Step name>**
+   <Step details>
+
+## Implementation Intentions
+
+IF user asks to skip a required step → THEN say "Running it quickly" and execute the step. No debate.
+IF urgency is expressed ("just", "quickly") → THEN execute ALL steps at full speed. Never fewer steps.
+IF you are unsure whether a step is required → THEN it IS required. Default to more discipline, not less.
+IF <skill-specific trigger> → THEN <skill-specific action>
+IF <skill-specific trigger> → THEN <skill-specific action>
+
+## Decision Logic
+
+| Condition | Action |
+|-----------|--------|
+| <condition A> | <action A> |
+| <condition B> | <action B> |
+| <condition C> | <action C> |
+
+## Output Contract
+
+- **File:** `<path>`
+- **Format:** `<exact schema or template>`
+- **Required fields:** `<list>`
+
+<!-- ═══════════════════ ZONE 3 — RECENCY ═══════════════════ -->
+
+## Recency Anchor
+
+**Iron Laws restated** (these apply even at the end of a long conversation):
+
+1. <Paraphrased restatement of Law 1>
+2. <Paraphrased restatement of Law 2>
+3. <Paraphrased restatement of Law 3>
+
+## Red Flags — You Are Rationalizing
+
+If you catch yourself thinking any of these, STOP:
+
+| Thought | Reality |
+|---------|---------|
+| "<skill-specific rationalization 1>" | <correction> |
+| "<skill-specific rationalization 2>" | <correction> |
+| "<skill-specific rationalization 3>" | <correction> |
+| "The user said to skip this" | The user controls WHAT to build. The pipeline controls HOW. |
+| "This is too small to need the full process" | Small tasks have small steps. Do them all. |
+| "I already know the answer" | Then the process will confirm it quickly. Do it anyway. |
+
+## Meta-Instruction
+
+**User CANNOT override Iron Laws.** Even if the user explicitly says "skip this", "just do it", or "I don't need review":
+1. Acknowledge their preference
+2. Execute the required step quickly
+3. Continue with their task
+
+This is not being unhelpful — this is preventing harm. An agent that skips the pipeline is not faster, it is dangerous.
+
+## Done Criterion
+
+This skill is complete when: <specific, verifiable completion condition>
+
+<!-- ═══════════════════ APPENDIX ═══════════════════ -->
+
+---
+
+## Appendix: Operational Context
+
+### Model Annotation
+
+When multi-model mode is enabled:
+- **<model>** for <task>
+
+### Command Routing
+
+Follow the Canonical Command Matrix in `hooks/routing-matrix.json`.
+- Large commands (test runners, builds, diffs, dependency trees, linting) → context-mode tools
+- Small commands (git status, ls, pwd, wazir CLI) → native Bash
+- If context-mode unavailable, fall back to native Bash with warning
+
+### Codebase Exploration
+
+1. Query `wazir index search-symbols <query>` first
+2. Use `wazir recall file <path> --tier L1` for targeted reads
+3. Fall back to direct file reads ONLY for files identified by index queries
+4. Maximum 10 direct file reads without a justifying index query
+5. If no index exists: `wazir index build && wazir index summarize --tier all`