wicked-garden 12.28.0 → 12.28.1

package/.claude-plugin/marketplace.json

@@ -7,9 +7,9 @@
     {
       "name": "wicked-garden",
       "source": "./",
-      "description": "Your coding agent already plans and swarms \u2014 Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, real multi-model second opinions. Don't fight the harness; fill its gaps. The gate needs wicked-vault + wicked-loom; testing/brain/bus are opt-in layers.",
-      "version": "12.28.0"
+      "description": "Your coding agent already plans and swarms \u2014 Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, real multi-model second opinions. Don't fight the harness; fill its gaps. The gate needs wicked-vault (auto-installed via wicked-testing) and re-derives through garden's built-in engine; testing/brain/bus are opt-in layers.",
+      "version": "12.28.1"
     }
   ],
-  "version": "12.28.0"
+  "version": "12.28.1"
 }

package/.claude-plugin/plugin.json

@@ -1,10 +1,10 @@
 {
   "name": "wicked-garden",
-  "version": "12.28.0",
-  "description": "Your coding agent's harness already plans and swarms; Wicked Garden is the curated toolkit for what it can't do alone. Proof, not claims \u2014 gates re-derive 'done' through the internal loom engine (fail-closed, independently attested via wicked-vault), so a self-asserted pass can't lie its way green. Relationships grep can't see \u2014 codegraph + injected bus/dispatch/capability edges power blast-radius & lineage. Plus deterministic multi-file refactor (wicked-patch), cross-session memory (wicked-brain), real multi-model second opinions (jam:council), portable expertise as on-demand skill-refs, and evidence-gated testing (wicked-testing). It reads the shape of your work to apply the right rigor \u2014 steering, not a pipeline \u2014 then gets out of the way. The compiler stamps a self-contained evidence gate into any repo that runs with no wicked-garden installed. Don't fight the harness; fill its gaps. The evidence gate requires wicked-vault (installed automatically via wicked-testing); wicked-testing, wicked-brain, wicked-understanding, and wicked-bus are opt-in layers \u2014 install what you'll use, the toolkit works without the rest.",
-  "wicked_testing_version": "^0.7.0",
+  "version": "12.28.1",
+  "description": "Your coding agent's harness already plans and swarms; Wicked Garden is the curated toolkit for what it can't do alone. Proof, not claims \u2014 gates re-derive 'done' through the internal loom engine (fail-closed, independently attested via wicked-vault), so a self-asserted pass can't lie its way green. Relationships grep can't see \u2014 codegraph + injected bus/dispatch/capability edges power blast-radius & lineage. Plus deterministic multi-file refactor (wicked-patch), cross-session memory (wicked-brain), real multi-model second opinions (jam:council), portable expertise as on-demand skill-refs, and evidence-gated testing (wicked-testing). It reads the shape of your work to apply the right rigor \u2014 steering, not a pipeline \u2014 then gets out of the way. The compiler stamps a self-contained evidence gate into any repo that runs with no wicked-garden installed. Don't fight the harness; fill its gaps. The evidence gate requires wicked-vault (installed automatically via wicked-testing); wicked-testing, wicked-brain, and wicked-bus are opt-in layers \u2014 install what you'll use, the toolkit works without the rest.",
+  "wicked_testing_version": "^0.8.0",
   "wicked_vault_version": "^0.4.0",
-  "wicked_brain_version": "^0.15.0",
+  "wicked_brain_version": "^0.18.0",
   "wicked_bus_version": "^2.0.0",
   "author": {
     "name": "Mike Parcewski",

package/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 ## [Unreleased]
 
+## [12.28.1] — 2026-07-12
+
+### Fixed
+- **Peer version floors unblocked (P0).** `wicked_testing_version` `^0.7.0`→`^0.8.0` and `wicked_brain_version` `^0.15.0`→`^0.18.0`. The testing floor is *enforced fail-closed* and its caret excluded the published wicked-testing 0.8.0, hard-blocking garden's produces-gate for anyone on testing 0.8.0.
+- Dropped the dead `npx wicked-vault-install` command from user docs (vault installs automatically via wicked-testing) and stale standalone references to absorbed/archived products (wicked-loom is now internal to garden; wicked-understanding is archived) from the plugin description + docs.
+
 ## [12.28.0] — 2026-07-12
 
 ### Changed

package/CONTRIBUTING.md

@@ -48,7 +48,7 @@ setup wizard output, or the README. It is not a supported production configurati
 
 **For offline CI and dev environments without the wicked-vault peer only.**
 
-wicked-vault (npm, ≥0.3, install `npx wicked-vault-install`) is a required peer
+wicked-vault (npm, ≥0.3, install `npx wicked-testing install`) is a required peer
 alongside wicked-loom/bus/brain/testing. The garden's produces-gates re-derive
 evidence through **wicked-loom** — `scripts/qe/vault_gate.py` shells `wicked-loom
 gate`, which in turn shells `wicked-vault cross-check`, so the vault is the backend

package/README.md

@@ -61,7 +61,7 @@ No universal pipeline to obey. A hook reads each prompt's *shape* and that decid
 **One required peer, the rest opt-in.** The evidence gate is the floor we won't fake, so it needs one external peer — setup blocks without it:
 
 ```bash
-npx wicked-vault-install       # the honest-evidence backend the gate re-derives against (≥ 0.4.0)
+npx wicked-testing install     # installs wicked-vault (≥ 0.4.0), the honest-evidence backend the gate re-derives against
 ```
 
 > The gate/resolve engine (formerly the separate `wicked-loom` package) is now **absorbed in-package** as of v12.27.0 (`scripts/loom/`) — nothing extra to install. The gate re-hashes recorded evidence and re-runs its verifier through that engine; a false "tests pass" is **rejected**, a missing backend **fails closed**.
@@ -71,7 +71,6 @@ The rest of the kit is **opt-in layers** — add what you want, skip the rest an
 ```bash
 npx wicked-testing install     # the acceptance-testing layer (author ≠ executor ≠ reviewer)
 /plugin install wicked-brain   # the memory layer — cross-session recall + cited search (the "what")
-npx skills add mikeparcewski/wicked-understanding --all   # the repo-playbooks layer — "how to work in THIS repo" (the "how"; pairs with brain)
 /plugin install wicked-bus     # the audit-trail layer (fire-and-forget; fail-open without it)
 ```
 

package/docs/getting-started.md

@@ -18,17 +18,15 @@ No API keys, no external services, no cloud — everything runs locally.
 
 ### Required peer plugins
 
-Five companion plugins are **required** — `/wicked-garden-core setup` verifies all five and blocks until they are present. They are required at install but resilient at runtime (the garden degrades gracefully if one goes missing mid-session). Graceful degradation means a session continues where it's safe to; it never means a gate treats missing evidence as a pass — that path fails closed.
+Companion peers are set up alongside the toolkit — `/wicked-garden-core setup` verifies them and **blocks only on the one the evidence gate needs** (the wicked-vault backend, installed via wicked-testing). They are required at install but resilient at runtime (the garden degrades gracefully if one goes missing mid-session). Graceful degradation means a session continues where it's safe to; it never means a gate treats missing evidence as a pass — that path fails closed.
 
 ```bash
-npx wicked-testing install    # wicked-testing — evidence-gated testing
-npx wicked-vault-install      # wicked-vault — vault-backed evidence (loom re-runs the verifier against it)
-npx wicked-loom               # wicked-loom — the gate engine (re-derives "done" via the vault)
+npx wicked-testing install    # wicked-testing — evidence-gated testing; bundles the wicked-vault evidence backend the gate re-derives against
 /plugin install wicked-brain  # wicked-brain — cross-session memory + search
 /plugin install wicked-bus    # wicked-bus — event bridge
 ```
 
-`wicked-testing`, `wicked-vault`, and `wicked-loom` install locally via npm; `wicked-brain` and `wicked-bus` install as Claude Code plugins. Gates re-derive "done" through wicked-loom — which re-runs the verifier via wicked-vault — and fail closed if loom is unavailable; a claim is never self-asserted.
+The gate/resolve engine (**wicked-loom**) ships inside wicked-garden (`scripts/loom/`) — nothing to install for it. `wicked-testing` (which bundles `wicked-vault`) installs locally via npm; `wicked-brain` and `wicked-bus` install as Claude Code plugins. Gates re-derive "done" through that built-in loom engine — which re-runs the verifier via wicked-vault — and fail closed if the vault is unavailable; a claim is never self-asserted.
 
 ### Show the active mode (status line)
 
@@ -203,6 +201,6 @@ Your data is stored locally in `~/.something-wicked/wicked-garden/` as JSON file
 ## Next Steps
 
 - [Archetypes](v11/archetypes.md) — the 9 work-shapes: phases, produces, human-in-the-loop (HITL) discipline, cost bands
-- [Required Peers](required-peers.md) — wicked-testing, wicked-vault, wicked-brain, wicked-bus, wicked-loom
+- [Required Peers](required-peers.md) — wicked-testing, wicked-vault, wicked-brain, wicked-bus
 - [Compiler](compiler.md) — emit a standalone, vault-backed build gate into any repo
 - [Domains](domains.md) — browse the domain skills and their actions

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "wicked-garden",
-  "version": "12.28.0",
+  "version": "12.28.1",
   "type": "module",
   "description": "Curated toolkit for what AI coding agents can't do alone: evidence-gated work, codegraph relationships, deterministic multi-file refactor, multi-model second opinions, and cross-session memory. Installs as a Claude Code plugin.",
   "keywords": [
@@ -33,6 +33,7 @@
   },
   "main": "install.mjs",
   "scripts": {
+    "version": "node .github/scripts/sync-plugin-version.mjs && git add .claude-plugin/plugin.json .claude-plugin/marketplace.json",
     "prepublishOnly": "node install.mjs --version"
   },
   "engines": {

package/skills/core/refs/install.md

@@ -34,7 +34,7 @@ peer-resolution engine is built into wicked-garden (scripts/loom/) — nothing t
 for loom. Run vault install and show a progress line:
 
 ```bash
-npx wicked-vault-install
+npx wicked-testing install
 ```
 
 On failure: display the raw error and note "you can install wicked-vault manually via `npx wicked-testing install` and re-run the wicked-garden-core skill's `install` action to retry." Continue to step 3 regardless — step 5 will surface what's still missing.

package/skills/core/refs/setup.md

@@ -120,7 +120,7 @@ PY
 npx wicked-vault --version 2>/dev/null || echo "MISSING"
 ```
 
-- `MISSING` → blocking. wicked-vault is the evidence backend every archetype gate re-derives against — without it, "done" can only be self-asserted. Show "wicked-vault is not installed. wicked-garden requires it as a peer (sibling to wicked-bus / wicked-brain / wicked-testing)." **INTERACTIVE mode**: AskUserQuestion header "wicked-vault Required", options "Install now (Required)" = "Run: npx wicked-vault-install" / "Exit setup" = "Cancel — I'll install manually and re-run". **PLAIN_TEXT mode**: present numbered options and STOP. If install: run `npx wicked-vault-install` (installs the cross-CLI skills) and confirm the CLI resolves with `npx wicked-vault --version`. On failure, show stderr and exit with manual instructions (`npm i -g wicked-vault`). If exit: "Run `npx wicked-vault-install` then restart by invoking the wicked-garden-core skill's `setup` action."
+- `MISSING` → blocking. wicked-vault is the evidence backend every archetype gate re-derives against — without it, "done" can only be self-asserted. Show "wicked-vault is not installed. wicked-garden requires it as a peer (sibling to wicked-bus / wicked-brain / wicked-testing)." **INTERACTIVE mode**: AskUserQuestion header "wicked-vault Required", options "Install now (Required)" = "Run: npx wicked-testing install" / "Exit setup" = "Cancel — I'll install manually and re-run". **PLAIN_TEXT mode**: present numbered options and STOP. If install: run `npx wicked-testing install` (installs the cross-CLI skills) and confirm the CLI resolves with `npx wicked-vault --version`. On failure, show stderr and exit with manual instructions (`npm i -g wicked-vault`). If exit: "Run `npx wicked-testing install` then restart by invoking the wicked-garden-core skill's `setup` action."
 - Version string (e.g. `0.3.0`) → show "wicked-vault {version} — ready." Then verify the garden can resolve it for gating: `sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/qe/vault_gate.py" resolve` should report `resolvable: true`. If `installed: false` (resolving only via npx), suggest `npm i -g wicked-vault` for faster gate latency — recommended, not a hard block.
 
 ### 2.7 Verify wicked-bus (Recommended — audit-trail layer)

package/.wicked/worktrees/aaedbab8-9152-44ba-9d87-4ce25d0d4afd/.claude-plugin/archetypes.json

@@ -1,166 +0,0 @@
-{
-  "$schema_version": "v11.0.0",
-  "$comment": "Work-shape archetypes (v11). Each archetype is a complete unit with its own phases, produces, HITL discipline, and cost band. No universal pipeline. Detector classifies a prompt into one or more archetypes; steering emits per-archetype directives. v6.3 archetype detection (code-repo / docs-only / etc.) was about target *kind*; this is about work *shape*.",
-  "archetypes": {
-    "triage": {
-      "description": "Classify the work and route to the right archetype. The entry shape for any prompt that hasn't been routed yet.",
-      "phases": ["classify"],
-      "produces": ["routing-decision"],
-      "hitl": "none",
-      "cost_band": "negligible",
-      "maturity": "production",
-      "signals": {
-        "always_on": true
-      },
-      "next_archetypes": "any",
-      "min_complexity": 0
-    },
-    "explore": {
-      "description": "Diverge then converge on options when the problem space is open. Produces an option set or hypothesis, not a decision.",
-      "phases": ["frame", "diverge", "converge"],
-      "produces": ["option-set", "hypothesis"],
-      "hitl": "continuous",
-      "cost_band": "low",
-      "maturity": "research",
-      "signals": {
-        "phrases": ["what should we", "how might we", "explore options", "options for", "should we", "brainstorm", "approaches to", "not sure how", "ideas for", "could we"],
-        "novelty_high": true,
-        "ambiguity_high": true
-      },
-      "next_archetypes": ["specify", "decide", "build"],
-      "min_complexity": 0
-    },
-    "specify": {
-      "description": "Turn an ask into structured, testable acceptance criteria.",
-      "phases": ["elicit", "structure", "validate"],
-      "produces": ["smart-acceptance-criteria"],
-      "hitl": "discrete:validate-gate",
-      "cost_band": "low",
-      "maturity": "piloted",
-      "signals": {
-        "spec_ambiguity_high": true,
-        "phrases": ["acceptance criteria", "requirements", "user story", "spec"],
-        "scope_unclear": true
-      },
-      "next_archetypes": ["build", "decide"],
-      "min_complexity": 0
-    },
-    "decide": {
-      "description": "Pick between options with an ADR-shaped artifact. Used when 2+ paths are viable and the choice is load-bearing.",
-      "phases": ["brief", "options", "score", "record"],
-      "produces": ["adr", "decision-artifact"],
-      "hitl": "discrete:select-gate",
-      "cost_band": "medium",
-      "maturity": "piloted",
-      "signals": {
-        "phrases": ["should we use", "x or y", "adr", "decision", "trade-off", "pick a", "pick between", "choose between", "which database", "which queue", "which framework"],
-        "multiple_viable_options": true,
-        "reversibility_medium_or_low": true
-      },
-      "next_archetypes": ["build", "migrate", "ship"],
-      "min_complexity": 2
-    },
-    "ship": {
-      "description": "Roll out an already-built change with progressively widening blast radius.",
-      "phases": ["canary", "ramp", "full", "soak"],
-      "produces": ["rollout-verdict", "slo-snapshot"],
-      "hitl": "discrete:ramp-gates",
-      "cost_band": "medium",
-      "maturity": "piloted",
-      "signals": {
-        "phrases": ["deploy", "release", "rollout", "roll out", "ramp", "ship to", "ship the"],
-        "blast_radius_high": true,
-        "post_build": true
-      },
-      "next_archetypes": ["review"],
-      "min_complexity": 2
-    },
-    "review": {
-      "description": "Independent assessment of an artifact (code / spec / design / incident response). Hard verdict at the end — APPROVE / CONDITIONAL / REJECT — with a remediation list when not APPROVE.",
-      "phases": ["scope", "assess", "findings", "remediate-or-accept"],
-      "produces": ["verdict", "remediation-list"],
-      "hitl": "hard:final-verdict",
-      "cost_band": "medium",
-      "maturity": "piloted",
-      "signals": {
-        "phrases": ["review", "audit", "evaluate", "verdict", "code review", "design review", "take a look at", "look at this pr", "look at the pr"],
-        "independent_assessment_needed": true
-      },
-      "next_archetypes": ["build", "migrate", "ship", "incident"],
-      "min_complexity": 1
-    },
-    "incident": {
-      "description": "Live-fire response to a production failure. Mitigate first, RCA second, follow-up third.",
-      "phases": ["triage", "investigate", "mitigate", "resolve", "followup"],
-      "produces": ["mitigation", "rca", "followup-list"],
-      "hitl": "hard:mitigate",
-      "cost_band": "variable",
-      "maturity": "production",
-      "signals": {
-        "phrases": ["down", "outage", "incident", "broken in prod", "production is broken", "post-mortem", "postmortem", "5xx", "error rate", "memory leak", "paging us", "paged", "oncall paged"],
-        "production_impact": true
-      },
-      "next_archetypes": ["build", "migrate", "review"],
-      "min_complexity": 0
-    },
-    "build": {
-      "description": "Implement and ship a new feature or fix. The most common archetype. Test rigor scales with complexity / blast_radius / novelty.",
-      "phases": ["plan", "implement", "test", "review"],
-      "produces": ["shipped-code", "test-report"],
-      "hitl": "discrete:review-gate",
-      "cost_band": "high",
-      "maturity": "research",
-      "signals": {
-        "phrases": ["implement", "add", "build", "feature", "fix", "create", "refactor", "patch", "wire up", "wire in", "hook up", "update the"],
-        "code_change": true
-      },
-      "next_archetypes": ["ship", "review"],
-      "min_complexity": 0,
-      "test_rigor_scales_with": ["complexity", "blast_radius", "novelty"]
-    },
-    "migrate": {
-      "description": "Shape change that requires forward + rollback proof. Schema migrations, data backfills, breaking-API rollouts. Expand-contract pattern.",
-      "phases": ["plan", "expand", "backfill", "cutover", "contract"],
-      "produces": ["shape-change", "rollback-proof"],
-      "hitl": "hard:cutover-gate",
-      "cost_band": "high",
-      "maturity": "research",
-      "signals": {
-        "phrases": ["migration", "migrate", "schema change", "backfill", "cutover", "rename column", "rename the", "drop column", "breaking change", "expand the", "retire the", "take the", "column out"],
-        "state_complexity_high": true,
-        "reversibility_low": true
-      },
-      "next_archetypes": ["ship", "review"],
-      "min_complexity": 2
-    },
-    "modernize": {
-      "description": "Port a legacy codebase to a new stack: discover the estate, extract the business rules, blueprint the target, transform with leverage, prove parity, cut over. The factory phase shape, run as a GENERAL modernization steering shape that composes the general code-modernization skills — NOT in-place shape change (that is migrate).",
-      "phases": ["discover", "extract", "blueprint", "transform", "parity", "cutover"],
-      "produces": ["modernization-blueprint", "parity-proof"],
-      "hitl": "hard:cutover-gate",
-      "cost_band": "high",
-      "maturity": "research",
-      "signals": {
-        "phrases": ["modernize", "modernization", "port from", "port the", "rewrite the legacy", "legacy codebase", "lift and shift", "cobol", "angularjs", ".net framework", "rehost", "replatform", "new stack", "to a new stack"],
-        "novelty_high": true,
-        "state_complexity_high": true,
-        "reversibility_low": true
-      },
-      "next_archetypes": ["review", "ship"],
-      "min_complexity": 2
-    }
-  },
-  "$detector_notes": [
-    "Archetypes are NOT mutually exclusive. A schema-changing feature is build + migrate. A risky deploy is ship + review.",
-    "triage is the entry archetype — it always fires first to classify.",
-    "Signals come from the prompt text + the propose-process 9-factor scoring. The detector is allowed to be confident (>0.7) on a single archetype OR return a small set.",
-    "When no archetype scores > 0.5, return ['triage'] and ask a clarifying question.",
-    "The phase shapes are EACH archetype's own. There is no universal pipeline. A 'build' that needs a 'specify' first chains: specify -> build, NOT specify-as-build's-clarify-phase."
-  ],
-  "$hitl_levels": {
-    "none": "No human checkpoint. Fires and is done.",
-    "continuous": "Human is a participant throughout. Brainstorm-style.",
-    "discrete:<gate>": "Hard stop at the named gate; human approves before proceeding.",
-    "hard:<gate>": "Same as discrete but the gate cannot be auto-approved or bypassed without explicit override audit log."
-  }
-}

package/.wicked/worktrees/aaedbab8-9152-44ba-9d87-4ce25d0d4afd/.claude-plugin/components.json

@@ -1,75 +0,0 @@
-{
-  "_comment": "Issue #328: Component declarations for wicked-garden plugin. Claude Code plugin.json does not yet support inline component declarations beyond path overrides. This manifest documents all components for tooling, validation, and future schema support. When plugin.json gains a 'components' field, merge this data into plugin.json.",
-  "domains": ["agentic", "crew", "data", "engineering", "jam", "mem", "persona", "platform", "product", "qe", "search", "smaht"],
-  "summary": {
-    "commands": 77,
-    "agents": 23,
-    "skills": 67,
-    "hooks": 13,
-    "specialists": 7
-  },
-  "agents_by_domain": {
-    "agentic": ["architect", "performance-analyst", "safety-reviewer"],
-    "crew": ["implementer", "researcher", "reviewer"],
-    "data": ["data-engineer"],
-    "engineering": ["api-documentarian", "migration-engineer", "solution-architect"],
-    "jam": ["brainstorm-facilitator", "council", "quick-facilitator"],
-    "persona": ["persona-agent"],
-    "platform": ["compliance-officer", "privacy-expert", "security-engineer"],
-    "product": ["a11y-expert", "requirements-analyst", "ui-reviewer", "ux-designer", "value-strategist"],
-    "qe": ["semantic-reviewer"]
-  },
-  "skills_by_domain": {
-    "agentic": ["agentic-patterns", "context-engineering", "frameworks", "review-methodology", "trust-and-safety"],
-    "archetype": ["archetype"],
-    "classify": ["classify"],
-    "data": ["analysis", "data", "ml", "pipeline"],
-    "deliberate": ["deliberate"],
-    "engineering": ["architecture", "backend", "debugging", "docs", "engineering", "frontend", "integration", "large-scale-migration", "patch", "system-design", "unit-test-quality"],
-    "ground": ["ground"],
-    "integration-discovery": ["integration-discovery"],
-    "jam": ["jam"],
-    "multi-model": ["multi-model"],
-    "persona": ["persona"],
-    "platform": ["audit", "compliance", "errors", "gate-benchmark-rebaseline", "gh-cli", "github-actions", "gitlab-ci", "glab-cli", "health", "observability", "prereq-doctor"],
-    "product": ["acceptance-criteria", "accessibility", "analyze", "imagery", "listen", "mockup", "requirements-analysis", "requirements-graph", "requirements-migrate", "requirements-navigate", "screenshot", "strategy", "synthesize", "ux-review", "visual-review"],
-    "runtime-exec": ["runtime-exec"],
-    "search": ["codebase-narrator"],
-    "smaht": ["discovery", "intent", "propose-skills", "smaht"],
-    "swarm": ["swarm"],
-    "wickedizer": ["wickedizer"],
-    "workflow": ["workflow"],
-    "worktrees": ["worktrees"]
-  },
-  "hooks": ["SessionStart", "UserPromptSubmit", "PreToolUse", "PostToolUse", "PostToolUseFailure", "TaskCompleted", "SubagentStart", "SubagentStop", "PermissionRequest", "Notification", "PreCompact", "Stop", "SessionEnd"],
-  "specialists": [
-    {
-      "name": "engineering",
-      "role": "engineering"
-    },
-    {
-      "name": "platform",
-      "role": "devsecops"
-    },
-    {
-      "name": "product",
-      "role": "product"
-    },
-    {
-      "name": "qe",
-      "role": "quality-engineering"
-    },
-    {
-      "name": "data",
-      "role": "data-engineering"
-    },
-    {
-      "name": "jam",
-      "role": "brainstorming"
-    },
-    {
-      "name": "agentic",
-      "role": "agentic-architecture"
-    }
-  ]
-}

package/.wicked/worktrees/aaedbab8-9152-44ba-9d87-4ce25d0d4afd/.claude-plugin/marketplace.json

@@ -1,15 +0,0 @@
-{
-  "name": "wicked-garden",
-  "owner": {
-    "name": "Mike Parcewski"
-  },
-  "plugins": [
-    {
-      "name": "wicked-garden",
-      "source": "./",
-      "description": "Your coding agent already plans and swarms \u2014 Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, real multi-model second opinions. Don't fight the harness; fill its gaps. The gate needs wicked-vault + wicked-loom; testing/brain/bus are opt-in layers.",
-      "version": "12.27.0"
-    }
-  ],
-  "version": "12.27.0"
-}

package/.wicked/worktrees/aaedbab8-9152-44ba-9d87-4ce25d0d4afd/.claude-plugin/plugin.json

@@ -1,175 +0,0 @@
-{
-  "name": "wicked-garden",
-  "version": "12.27.0",
-  "description": "Your coding agent's harness already plans and swarms; Wicked Garden is the curated toolkit for what it can't do alone. Proof, not claims \u2014 gates re-derive 'done' through the internal loom engine (fail-closed, independently attested via wicked-vault), so a self-asserted pass can't lie its way green. Relationships grep can't see \u2014 codegraph + injected bus/dispatch/capability edges power blast-radius & lineage. Plus deterministic multi-file refactor (wicked-patch), cross-session memory (wicked-brain), real multi-model second opinions (jam:council), portable expertise as on-demand skill-refs, and evidence-gated testing (wicked-testing). It reads the shape of your work to apply the right rigor \u2014 steering, not a pipeline \u2014 then gets out of the way. The compiler stamps a self-contained evidence gate into any repo that runs with no wicked-garden installed. Don't fight the harness; fill its gaps. The evidence gate requires wicked-vault (installed automatically via wicked-testing); wicked-testing, wicked-brain, wicked-understanding, and wicked-bus are opt-in layers \u2014 install what you'll use, the toolkit works without the rest.",
-  "wicked_testing_version": "^0.7.0",
-  "wicked_vault_version": "^0.4.0",
-  "wicked_brain_version": "^0.15.0",
-  "wicked_bus_version": "^2.0.0",
-  "author": {
-    "name": "Mike Parcewski",
-    "email": "mike.parcewski@gmail.com"
-  },
-  "repository": "https://github.com/mikeparcewski/wicked-garden",
-  "license": "MIT",
-  "keywords": [
-    "coding-agent",
-    "harness",
-    "evidence-gate",
-    "code-graph",
-    "code-intelligence",
-    "refactoring",
-    "testing",
-    "memory",
-    "multi-model",
-    "agentic"
-  ],
-  "_future_userConfig": {
-    "_comment": "Issue #335, #346: userConfig is not yet supported in plugin.json schema. When Claude Code adds userConfig support, move these fields out of _future_userConfig and into a top-level userConfig block. This would replace the manual /wicked-garden:setup flow.",
-    "storage_mode": {
-      "type": "string",
-      "default": "local",
-      "enum": [
-        "local",
-        "sqlite"
-      ],
-      "description": "Where to persist plugin state: local JSON files or SQLite with FTS5"
-    },
-    "onboarding_preference": {
-      "type": "string",
-      "default": "auto",
-      "enum": [
-        "auto",
-        "manual",
-        "skip"
-      ],
-      "description": "How to handle first-run onboarding: auto-detect and configure, prompt user, or skip"
-    },
-    "log_level": {
-      "type": "string",
-      "default": "info",
-      "enum": [
-        "debug",
-        "info",
-        "warn",
-        "error"
-      ],
-      "description": "Verbosity level for operational logs"
-    },
-    "jira_api_token": {
-      "_comment": "Issue #346: When userConfig supports sensitive:true, this field should use it to store the token in the OS keychain rather than plaintext.",
-      "type": "string",
-      "sensitive": true,
-      "description": "Jira API token for issue sync (stored in OS keychain when sensitive flag is supported)"
-    },
-    "linear_api_token": {
-      "type": "string",
-      "sensitive": true,
-      "description": "Linear API token for issue sync (stored in OS keychain when sensitive flag is supported)"
-    },
-    "slack_webhook_url": {
-      "type": "string",
-      "sensitive": true,
-      "description": "Slack webhook URL for notifications (stored in OS keychain when sensitive flag is supported)"
-    }
-  },
-  "_future_channels": {
-    "_comment": "Issue #348: channels field is not yet supported in plugin.json schema. When Claude Code adds channel support, move these declarations to a top-level channels block. Channels formalize external service connections with discovery and auth.",
-    "jira": {
-      "type": "rest-api",
-      "description": "Jira issue tracker for bi-directional sync with native Claude Code tasks",
-      "auth": "oauth2",
-      "capabilities": [
-        "read-issues",
-        "write-issues",
-        "webhooks"
-      ]
-    },
-    "linear": {
-      "type": "rest-api",
-      "description": "Linear issue tracker for bi-directional sync with native Claude Code tasks",
-      "auth": "api-key",
-      "capabilities": [
-        "read-issues",
-        "write-issues",
-        "webhooks"
-      ]
-    },
-    "slack": {
-      "type": "webhook",
-      "description": "Slack notifications for crew phase completions and gate failures",
-      "auth": "webhook-url",
-      "capabilities": [
-        "send-messages"
-      ]
-    },
-    "github": {
-      "type": "rest-api",
-      "description": "GitHub for issue triage, PR creation, and CI status",
-      "auth": "token",
-      "capabilities": [
-        "read-issues",
-        "write-issues",
-        "read-prs",
-        "write-prs",
-        "read-checks"
-      ]
-    }
-  },
-  "_future_lspServers": {
-    "_comment": "Issue #347: lspServers field enables Language Server Protocol integration for richer code intelligence (go-to-definition, find-references, hover, diagnostics). When Claude Code adds lspServers support to plugins, investigate integrating with the search/CIQ domain for enhanced symbol resolution. Candidate servers: TypeScript (typescript-language-server), Python (pyright), Rust (rust-analyzer)."
-  },
-  "_future_outputStyles": {
-    "_comment": "Issue #338: outputStyles field is not yet supported as a configuration block in plugin.json. When Claude Code adds outputStyles support, move these declarations to a top-level outputStyles block or the ./styles/ directory. These define domain-specific formatting for different work modes.",
-    "crew-phase": {
-      "description": "Structured output for crew workflow phases with phase headers, gate results, and evidence summaries",
-      "format": "markdown",
-      "sections": [
-        "phase-header",
-        "deliverables",
-        "gate-result",
-        "evidence-summary"
-      ]
-    },
-    "engineering-review": {
-      "description": "Code review output with severity-tagged findings, file annotations, and action items",
-      "format": "markdown",
-      "sections": [
-        "summary",
-        "findings-by-severity",
-        "file-annotations",
-        "action-items"
-      ]
-    },
-    "delivery-report": {
-      "description": "Delivery metrics report with velocity charts, risk indicators, and forecast",
-      "format": "markdown",
-      "sections": [
-        "velocity",
-        "risks",
-        "forecast",
-        "recommendations"
-      ]
-    },
-    "qe-results": {
-      "description": "Test execution results with pass/fail counts, coverage, and failure analysis",
-      "format": "markdown",
-      "sections": [
-        "summary",
-        "results-table",
-        "failure-analysis",
-        "coverage"
-      ]
-    },
-    "brainstorm-synthesis": {
-      "description": "Jam session output with perspectives, convergence points, and actionable next steps",
-      "format": "markdown",
-      "sections": [
-        "perspectives",
-        "convergence",
-        "divergence",
-        "next-steps"
-      ]
-    }
-  }
-}

package/.wicked/worktrees/aaedbab8-9152-44ba-9d87-4ce25d0d4afd/.claude-plugin/specialist.json

@@ -1,48 +0,0 @@
-{
-  "plugin": "wicked-garden",
-  "_comment": "Specialist roles listed here are user-facing disciplines surfaced by crew for phase routing. Internal infrastructure domains (crew, mem, persona, smaht, search, jam) are NOT listed as specialists — they are invoked directly via their commands, not routed through the specialist dispatch system.",
-  "specialists": [
-    {
-      "name": "engineering",
-      "role": "engineering",
-      "description": "Full-stack engineering expertise spanning code quality, architecture, debugging, documentation, and implementation patterns.",
-      "enhances": ["design", "build", "review"]
-    },
-    {
-      "name": "platform",
-      "role": "devsecops",
-      "description": "DevSecOps specialist for security scanning, CI/CD pipelines, infrastructure, compliance, and incident response.",
-      "enhances": ["build", "review", "operate", "*"]
-    },
-    {
-      "name": "product",
-      "role": "product",
-      "description": "Product and design expertise spanning requirements, UX, customer voice, business strategy, accessibility, visual design review, mockups, and user flow analysis.",
-      "enhances": ["clarify", "design", "review"]
-    },
-    {
-      "name": "qe",
-      "role": "quality-engineering",
-      "description": "Quality Engineering specialist powered by wicked-testing peer plugin. Crew dispatches QE agents by their wicked-testing:* subagent_type names. Covers test strategy, scenario generation, testability review, semantic alignment, risk assessment, test automation, and shift-left quality practices. Requires wicked-testing ^0.2 installed.",
-      "enhances": ["clarify", "design", "qe", "build", "review", "operate", "*"]
-    },
-    {
-      "name": "data",
-      "role": "data-engineering",
-      "description": "Data engineering specialist providing expertise in pipeline design, data analysis, ML model guidance, and analytics architecture.",
-      "enhances": ["design", "build", "*"]
-    },
-    {
-      "name": "jam",
-      "role": "brainstorming",
-      "description": "Multi-perspective brainstorming with dynamic focus groups. Explores ideas through persona-based discussion rounds and synthesis.",
-      "enhances": ["clarify", "design", "build"]
-    },
-    {
-      "name": "agentic",
-      "role": "agentic-architecture",
-      "description": "Validates agentic application architecture, patterns, and design. Reviews agent implementations for safety, performance, and best practices.",
-      "enhances": ["design", "build", "review"]
-    }
-  ]
-}