@wazir-dev/cli 1.0.0

package/exports/hosts/claude/.claude/commands/design.md

@@ -0,0 +1,44 @@
+# design
+
+## Purpose
+
+Transform the approved spec into visual designs using open-pencil MCP tools.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- approved spec artifact
+- research artifacts
+- brand/style guidelines (if available)
+- author artifact
+
+## Primary Role
+
+- `designer`
+
+## Outputs
+
+- design artifact (`.fig` + exported code + tokens + screenshots)
+
+## Skip Conditions
+
+- project has no visual design needs
+- open-pencil MCP server is not available (proceed to planning with text-only design specs)
+
+## Approval Gate
+
+- explicit human approval required before design-review
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- design does not address spec requirements
+- missing exported code scaffolds or tokens

package/exports/hosts/claude/.claude/commands/discover.md

@@ -0,0 +1,37 @@
+# discover
+
+## Purpose
+
+Gather source-backed research that reduces planning and implementation risk.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- clarified scope
+- explicit research questions
+
+## Primary Role
+
+- `researcher`
+
+## Outputs
+
+- research artifact
+- cited findings
+
+## Approval Gate
+
+- no formal approval gate, but unsupported research cannot flow forward
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- missing citations

package/exports/hosts/claude/.claude/commands/execute.md

@@ -0,0 +1,48 @@
+# execute
+
+## Purpose
+
+Implement the approved plan in ordered, verified batches.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Pre-execution validation
+
+Before implementing, run these checks to ensure the repo is in a valid state:
+- `wazir validate manifest` — confirm the manifest schema is still valid
+- `wazir validate hooks` — confirm hook contracts are intact
+
+If either check fails:
+1. Surface the failure to the user
+2. Do NOT proceed with implementation until the issue is resolved
+
+## Inputs
+
+- approved implementation plan
+- current branch state
+- author artifact (i18n keys, seed data, notification templates)
+
+## Primary Role
+
+- `executor`
+
+## Outputs
+
+- code and docs changes
+- execution notes
+
+## Approval Gate
+
+- no new scope without explicit approval
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- plan drift

package/exports/hosts/claude/.claude/commands/learn.md

@@ -0,0 +1,38 @@
+# learn
+
+## Purpose
+
+Extract durable scoped learnings and experiments from the completed run.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- run artifacts
+- review findings
+- verification proof
+
+## Primary Role
+
+- `learner`
+
+## Outputs
+
+- proposed learning artifacts
+- experiment summaries
+
+## Approval Gate
+
+- accepted learnings require explicit review and scope tags
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- auto-applied learning drift

package/exports/hosts/claude/.claude/commands/plan-review.md

@@ -0,0 +1,42 @@
+# plan-review
+
+## Purpose
+
+Adversarially review the implementation plan before execution starts.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- draft implementation plan
+- approved spec
+
+## Primary Role
+
+- `reviewer`
+
+## Outputs
+
+- plan review findings
+- revised tasks when needed
+
+## Approval Gate
+
+- execution cannot begin until blocking findings are closed or explicitly accepted
+
+## Gate decision
+
+On approval: `wazir capture event --run <run-id> --event gate_approved --phase <phase-name>`
+On rejection: `wazir capture event --run <run-id> --event gate_rejected --phase <phase-name> --message "<reason>"`
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- sequence gaps survive review

package/exports/hosts/claude/.claude/commands/plan.md

@@ -0,0 +1,39 @@
+# plan
+
+## Purpose
+
+Translate the approved spec into an execution-grade implementation plan.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- approved spec
+- repo state
+- research artifacts
+- author artifact
+
+## Primary Role
+
+- `planner`
+
+## Outputs
+
+- implementation plan artifact
+- ordered tasks and verification steps
+
+## Approval Gate
+
+- explicit human approval required before execution
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- missing dependencies or unverifiable steps

package/exports/hosts/claude/.claude/commands/prepare-next.md

@@ -0,0 +1,37 @@
+# prepare-next
+
+## Purpose
+
+Prepare the next run or next execution slice without silently carrying stale context forward.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- current run summary
+- accepted learnings where explicitly enabled
+
+## Primary Role
+
+- `planner`
+
+## Outputs
+
+- next-step handoff
+- scoped context summary
+
+## Approval Gate
+
+- no implicit carry-forward of unapproved learnings
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- stale context treated as current truth

package/exports/hosts/claude/.claude/commands/review.md

@@ -0,0 +1,40 @@
+# review
+
+## Purpose
+
+Run adversarial review against the changed implementation and its evidence.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- changed files
+- verification proof
+- approved spec and plan
+- design artifact (when design phase was used)
+
+## Primary Role
+
+- `reviewer`
+
+## Outputs
+
+- findings
+- no-findings verdict when applicable
+
+## Approval Gate
+
+- unresolved blocking findings must stop completion
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- vague review or uncited findings
+- design-implementation mismatch not flagged (when design artifact exists)

package/exports/hosts/claude/.claude/commands/run-audit.md

@@ -0,0 +1,41 @@
+# run-audit
+
+## Purpose
+
+Perform a structured audit on the codebase, branch, or scoped paths and produce source-backed findings with severity, evidence, citations, and remediation.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- audit type (security, code-quality, architecture, performance, dependencies, custom)
+- audit scope (whole-project, branch, paths)
+- output mode (report, plan)
+
+## Primary Role
+
+- `researcher` (composed with audit-specific expertise from `audit-*` concerns)
+
+## Outputs
+
+- structured audit report with artifact metadata (report mode)
+- approved design doc in `docs/plans/` for `wz:writing-plans` handoff (plan mode)
+
+## Approval Gate
+
+- no finding without source-backed evidence (file path, line, snippet, citation)
+- no severity rating without justification
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- superficial findings without verification
+- unsupported claims or missing citations
+- skipping files within declared scope

package/exports/hosts/claude/.claude/commands/spec-challenge.md

@@ -0,0 +1,41 @@
+# spec-challenge
+
+## Purpose
+
+Adversarially test the proposed spec for contradictions, omissions, and fake completeness.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- draft spec
+- research artifacts
+
+## Primary Role
+
+- `reviewer`
+
+## Outputs
+
+- spec challenge findings
+
+## Approval Gate
+
+- spec cannot be treated as approved until challenge findings are resolved or explicitly accepted
+
+## Gate decision
+
+On approval: `wazir capture event --run <run-id> --event gate_approved --phase <phase-name>`
+On rejection: `wazir capture event --run <run-id> --event gate_rejected --phase <phase-name> --message "<reason>"`
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- rubber-stamp review

package/exports/hosts/claude/.claude/commands/specify.md

@@ -0,0 +1,38 @@
+# specify
+
+## Purpose
+
+Turn clarified scope and research into a measurable spec.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- clarification artifact
+- research artifacts
+
+## Primary Role
+
+- `specifier`
+
+## Outputs
+
+- spec artifact
+- acceptance criteria
+- assumptions and non-goals
+
+## Approval Gate
+
+- explicit human approval required before planning
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- acceptance criteria are not testable

package/exports/hosts/claude/.claude/commands/verify.md

@@ -0,0 +1,37 @@
+# verify
+
+## Purpose
+
+Produce deterministic proof for the claims made by execution.
+
+## Phase entry
+
+On entering this phase, run:
+`wazir capture event --run <run-id> --event phase_enter --phase <phase-name> --status in_progress`
+
+## Inputs
+
+- changed files
+- claimed outcomes
+- acceptance criteria
+
+## Primary Role
+
+- `verifier`
+
+## Outputs
+
+- verification proof artifact
+
+## Approval Gate
+
+- no completion claim without fresh proof
+
+## Phase exit
+
+On completing this phase, run:
+`wazir capture event --run <run-id> --event phase_exit --phase <phase-name> --status completed`
+
+## Failure Conditions
+
+- stale or partial verification

package/exports/hosts/claude/.claude/settings.json

@@ -0,0 +1,34 @@
+{
+  "hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Write|Edit",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "./hooks/protected-path-write-guard"
+          }
+        ]
+      }
+    ],
+    "SessionStart": [
+      {
+        "hooks": [
+          {
+            "type": "command",
+            "command": "./hooks/loop-cap-guard"
+          }
+        ]
+      },
+      {
+        "matcher": "startup|resume|clear|compact",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "./hooks/session-start"
+          }
+        ]
+      }
+    ]
+  }
+}

package/exports/hosts/claude/CLAUDE.md

@@ -0,0 +1,19 @@
+# Wazir for Claude
+
+This host package is generated from the canonical Wazir sources.
+
+## Canonical facts
+
+- project: Wazir
+- hosts: claude, codex, gemini, cursor
+- phases: clarify, discover, specify, spec_challenge, author, design, design_review, plan, plan_review, execute, verify, review, learn, prepare_next
+- roles: clarifier, researcher, specifier, content-author, designer, planner, executor, verifier, reviewer, learner
+- protected paths: input, roles, workflows, schemas, exports/hosts
+- state root default: ~/.wazir/projects/{project_slug}
+
+## Source of truth
+
+- `wazir.manifest.yaml`
+- `roles/*.md`
+- `workflows/*.md`
+- `hooks/definitions/*.yaml`

package/exports/hosts/claude/export.manifest.json

@@ -0,0 +1,38 @@
+{
+  "host": "claude",
+  "source_hashes": {
+    "wazir.manifest.yaml": "2a57e262d071283504311ead7daaa8444c0c1895243f048036f79c501dfa4356",
+    "roles/clarifier.md": "1135b98d548937b5009bd77d292ceee29f806acc9cae245e5e18ccb8be05173e",
+    "roles/content-author.md": "cc20b80bd70ab68b3239a9cf56bf1ffc2c06843d38afc6b190844b35a1d73c3e",
+    "roles/designer.md": "8a0a9846965aa03ac00d0d71cde58f529aa8fb79908a481cb8ca3ff09d5af8df",
+    "roles/executor.md": "0de6cff7284d0bdf5fd3995e6cd5f66ad9479853829e4586fa6f793a4cb189c7",
+    "roles/learner.md": "4a7e4635c0fbd2def794f017ca6ef5d47b6ea8dfaa5e85eeeecece61256f21a1",
+    "roles/planner.md": "a8714eceaeb73ee1f0bb22030a00243881326b77d88b34b0bdd01fcba567aa69",
+    "roles/researcher.md": "1f521aecbb65680dd7c9747483acf8c34febb2159300dd4a39dc11fa857a3782",
+    "roles/reviewer.md": "a5d3ca44b6be2c432df02d8ce7889bdadf954582f5559ae7f4e24b05806a766b",
+    "roles/specifier.md": "9eb2dbe5319a87b2332c7d13601dcb70ca1a655adf947f473687e268dcd92fea",
+    "roles/verifier.md": "ce8e200439190643a02ffe31f72e4b4ec319ccff9f0841cc3138eb517af5fed3",
+    "workflows/author.md": "81844443743540b21dd990fa93e59d8898216e498acde920e7007f8b81cef107",
+    "workflows/clarify.md": "06354f7624ba2819fe10678eee05676bc737473194f444f7f7a9b0be1a9f036f",
+    "workflows/design-review.md": "0e48be715e11c719dee6d207970544eaaa86d755c9813b358e2b6d79afaeff90",
+    "workflows/design.md": "c68e78d3d078f9bba7a7b58b2c606ee1a2dd7b15f16567abab31adf3925e261e",
+    "workflows/discover.md": "af21ff43a57cce34ce37fbc84e82961d24968a2b95692d081b1697011f559ad8",
+    "workflows/execute.md": "49f4b4b22e50aac17c501d4b2d81b51acb82a7938f4a89f076552686cff7795d",
+    "workflows/learn.md": "9b1955a00eb0ea47af08a3639c3cbc2f15b8eab6cc02127e829bc8c83b0a52b5",
+    "workflows/plan-review.md": "fe6193122f7026dd92ced8ee1ae07279520d34b22a242260419cbb1d8bf7f99e",
+    "workflows/plan.md": "6e3a376b85c30e1887b5f667d28b51635a4332e05f892e49a17ae710c6ef247b",
+    "workflows/prepare-next.md": "8769b692be6d9fddf3b3f36fee6888159fb9fb2a5c31360cdbc9402e0f43fc27",
+    "workflows/review.md": "aaeaf7ab4515e325084a77e75e18b325295f8d16799f71f8b7cd0ec67c52c0d6",
+    "workflows/run-audit.md": "e582f767967dc3fb6aaeb938ad1672e7bed18bf044055ce8c45c2114251b787a",
+    "workflows/spec-challenge.md": "69c4114579a210bfe1333e92c382c771a6507ce71c4aecdff40f50c1818e06e2",
+    "workflows/specify.md": "477f7b651112383ea681d31d1fcecba7f3107946829f5660afbce9d97b7af4bd",
+    "workflows/verify.md": "3a568f1b32f6ece7a3d1382b0c75700337016cbee5d24ad8229a57339dc0ee98",
+    "hooks/definitions/loop_cap_guard.yaml": "143ecda8f47dd299182dfe9cea2fb3f15a82a8d1f179b100a9e539d83f639b57",
+    "hooks/definitions/post_tool_capture.yaml": "a773cd6e18972dee8eef3b7cb06fd1d319a71de4588897cebfbe643f6781a3b2",
+    "hooks/definitions/pre_compact_summary.yaml": "daa0175d79f3e0127c5ce86a7a2f8df0be3f58b5c94fe749da715a17c7b2d04e",
+    "hooks/definitions/pre_tool_capture_route.yaml": "3c2663380ff3cd09f09de5b96bcf6123266fa74d8a03dfb2d6fbe40a43fb13cf",
+    "hooks/definitions/protected_path_write_guard.yaml": "6683d41778b823e2a4e606065597569aa04363f091e135e165de9732f1fc2171",
+    "hooks/definitions/session_start.yaml": "9383fcf1f8304c87e57726478a461706c0fc73dc62bcc4d8661f2eeffa43a82d",
+    "hooks/definitions/stop_handoff_harvest.yaml": "67a3c0a8bb7cb66b88e77dc79e748082e964d278c47935662c453922a846482b"
+  }
+}

package/exports/hosts/claude/host-package.json

@@ -0,0 +1,67 @@
+{
+  "host": "claude",
+  "sources": [
+    "wazir.manifest.yaml",
+    "roles/clarifier.md",
+    "roles/content-author.md",
+    "roles/designer.md",
+    "roles/executor.md",
+    "roles/learner.md",
+    "roles/planner.md",
+    "roles/researcher.md",
+    "roles/reviewer.md",
+    "roles/specifier.md",
+    "roles/verifier.md",
+    "workflows/author.md",
+    "workflows/clarify.md",
+    "workflows/design-review.md",
+    "workflows/design.md",
+    "workflows/discover.md",
+    "workflows/execute.md",
+    "workflows/learn.md",
+    "workflows/plan-review.md",
+    "workflows/plan.md",
+    "workflows/prepare-next.md",
+    "workflows/review.md",
+    "workflows/run-audit.md",
+    "workflows/spec-challenge.md",
+    "workflows/specify.md",
+    "workflows/verify.md",
+    "hooks/definitions/loop_cap_guard.yaml",
+    "hooks/definitions/post_tool_capture.yaml",
+    "hooks/definitions/pre_compact_summary.yaml",
+    "hooks/definitions/pre_tool_capture_route.yaml",
+    "hooks/definitions/protected_path_write_guard.yaml",
+    "hooks/definitions/session_start.yaml",
+    "hooks/definitions/stop_handoff_harvest.yaml"
+  ],
+  "files": [
+    ".claude/agents/clarifier.md",
+    ".claude/agents/content-author.md",
+    ".claude/agents/designer.md",
+    ".claude/agents/executor.md",
+    ".claude/agents/learner.md",
+    ".claude/agents/planner.md",
+    ".claude/agents/researcher.md",
+    ".claude/agents/reviewer.md",
+    ".claude/agents/specifier.md",
+    ".claude/agents/verifier.md",
+    ".claude/commands/author.md",
+    ".claude/commands/clarify.md",
+    ".claude/commands/design-review.md",
+    ".claude/commands/design.md",
+    ".claude/commands/discover.md",
+    ".claude/commands/execute.md",
+    ".claude/commands/learn.md",
+    ".claude/commands/plan-review.md",
+    ".claude/commands/plan.md",
+    ".claude/commands/prepare-next.md",
+    ".claude/commands/review.md",
+    ".claude/commands/run-audit.md",
+    ".claude/commands/spec-challenge.md",
+    ".claude/commands/specify.md",
+    ".claude/commands/verify.md",
+    ".claude/settings.json",
+    "CLAUDE.md"
+  ]
+}

package/exports/hosts/codex/AGENTS.md

@@ -0,0 +1,19 @@
+# Wazir for Codex
+
+This host package is generated from the canonical Wazir sources.
+
+## Canonical facts
+
+- project: Wazir
+- hosts: claude, codex, gemini, cursor
+- phases: clarify, discover, specify, spec_challenge, author, design, design_review, plan, plan_review, execute, verify, review, learn, prepare_next
+- roles: clarifier, researcher, specifier, content-author, designer, planner, executor, verifier, reviewer, learner
+- protected paths: input, roles, workflows, schemas, exports/hosts
+- state root default: ~/.wazir/projects/{project_slug}
+
+## Source of truth
+
+- `wazir.manifest.yaml`
+- `roles/*.md`
+- `workflows/*.md`
+- `hooks/definitions/*.yaml`