opencodekit 0.22.0 → 0.23.0

package/dist/template/.opencode/command/review-codebase.md

@@ -1,19 +1,11 @@
 ---
 description: Review code for quality, security, and compliance
-argument-hint: "[path|bead-id|pr-number|'all'] [--quick|--thorough]"
+argument-hint: "[path|pr-number|'all'] [--quick|--thorough]"
 agent: review
 ---
 
 # Review: $ARGUMENTS
 
-## Load Skills
-
-```typescript
-skill({ name: "beads" });
-skill({ name: "requesting-code-review" });
-skill({ name: "verification-gates" });
-```
-
 ## Determine Input Type
 
 | Input Type            | Detection                    | Action                     |
@@ -22,6 +14,7 @@ skill({ name: "verification-gates" });
 | Commit hash (40-char) | SHA pattern                  | `git show <hash>`          |
 | Branch name           | String, not matching above   | `git diff main...HEAD`     |
 | PR URL/number         | Contains "github.com" or "#" | `gh pr diff`               |
+| Current plan          | `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md`    | Review against spec        |
 
 ## Before You Review
 
@@ -54,22 +47,22 @@ For each changed file:
 - Read the full file to understand context
 - Don't rely on diff alone — code that looks wrong in isolation may be correct
 
-If bead provided, read `.beads/artifacts/$ID/prd.md` to review against spec.
+If plan exists, read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md` to review against spec.
 
 ## Phase 2: Determine Scope
 
 | Input                    | Scope                 | How to Get Code           |
 | ------------------------ | --------------------- | ------------------------- |
 | File/directory path      | That path only        | `read` or `glob` + `read` |
-| Bead ID (e.g., `br-123`) | Implementation vs PRD | `br show` then `git diff` |
+| Current plan             | Implementation vs spec | Read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md` then `git diff` |
 | PR number (e.g., `#45`)  | PR changes            | `gh pr diff 45`           |
 | `all` or empty           | Recent changes        | `git diff main...HEAD`    |
 
-If bead provided, read `.beads/artifacts/$ID/prd.md` to review against spec.
+If plan exists, read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md` to review against spec.
 
 ## Phase 3: Automated Checks
 
-Follow the [verification-gates](../skill/verification-gates/SKILL.md) skill protocol.
+Use the `verification-before-completion` skill to run gates.
 
 Check `package.json` scripts, `Makefile`, or `justfile` for project-specific commands first — prefer those over generic defaults.
 

package/dist/template/.opencode/command/ship.md

@@ -1,37 +1,20 @@
 ---
-description: Ship a bead - implement PRD tasks, verify, review, close
-argument-hint: "<bead-id>"
+description: Ship a plan - implement specs, verify, review, close
 agent: build
 ---
 
-# Ship: $ARGUMENTS
+# Ship
 
-Execute PRD tasks, verify each passes, run review, close the bead.
+Execute spec tasks, verify each passes, run review, mark complete.
 
-> **Workflow:** `/create` → **`/ship <id>`**
->
-> Bead MUST have `prd.md`. If not yet claimed, `/ship` auto-claims it.
+> **Workflow:** `/create` → **`/ship`**
 
 ## Load Skills
 
 ```typescript
-skill({ name: "beads" });
-skill({ name: "memory-grounding" });
-skill({ name: "workspace-setup" });
 skill({ name: "verification-before-completion" });
-skill({ name: "reflection-checkpoints" }); // Mid-point + completion checks during execution
-// For user-facing UI changes: skill({ name: "ux-quality-gates" });
-// If local web/browser verification needs stable URLs: skill({ name: "portless" });
 ```
 
-## Determine Input Type
-
-| Input Type | Detection                   | Action                     |
-| ---------- | --------------------------- | -------------------------- |
-| Bead ID    | Matches `br-xxx` or numeric | Ship that bead             |
-| Path       | File/directory path         | Not supported for ship     |
-| `all`      | Keyword                     | Ship all in_progress beads |
-
 ## Before You Ship
 
 - **Be certain**: Only ship if all tasks pass verification
@@ -39,7 +22,7 @@ skill({ name: "reflection-checkpoints" }); // Mid-point + completion checks duri
 - **Run the review**: Always spawn review agent before closing
 - **Verify goals**: Tasks completing ≠ goals achieved (use goal-backward verification)
 - **Commit before close**: Per-task commits required, don't ship without git history
-- **Ask before closing**: Never close bead without user confirmation
+- **Ask before closing**: Never close without user confirmation
 
 ## Available Tools
 
@@ -53,55 +36,37 @@ skill({ name: "reflection-checkpoints" }); // Mid-point + completion checks duri
 
 ## Phase 1: Guards
 
-### Memory Grounding
-
-Follow the [memory-grounding](../skill/memory-grounding/SKILL.md) skill protocol. Focus on: failed approaches to avoid repeating.
+### Memory Search
 
-### Bead Validation
+Search memory for: failed approaches to avoid repeating.
 
-```bash
-br show $ARGUMENTS
-```
+### Plan Validation
 
 Verify:
 
-- Bead is `in_progress` or unclaimed (auto-claim if needed)
-- `.beads/artifacts/$ARGUMENTS/prd.md` exists (if not, tell user to run `/create` first)
+- `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md` exists (if not, tell user to run `/create` first)
 
 Check what artifacts exist:
 
-Read `.beads/artifacts/$ARGUMENTS/` to check what artifacts exist.
-
-## Phase 1B: Auto-Claim (if not yet in_progress)
-
-If bead status is NOT `in_progress`, auto-claim it:
-
-```bash
-br update $ARGUMENTS --status in_progress
-```
-
-Then ask about workspace:
+Read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/` to check what artifacts exist (spec.md, plan.md, etc.).
 
 ### Workspace Setup
 
-Follow the [workspace-setup](../skill/workspace-setup/SKILL.md) skill protocol.
-
-**If bead is already `in_progress`:** Skip this phase entirely.
+Set up the workspace: create branch, install deps if needed.
 
 ## Phase 2: Route to Execution
 
-| Artifact exists | Action                                                   |
+| Artifact exists in `.opencode/artifacts/$(cat .opencode/artifacts/.active)/` | Action                                                   |
 | --------------- | -------------------------------------------------------- |
-| `plan.md`       | Load `executing-plans` skill, follow its batch process   |
-| `prd.json`      | Proceed to PRD task loop below                           |
-| Only `prd.md`   | Load `prd-task` skill to create `prd.json`, then proceed |
+| `plan.md`       | Parse plan header + dependency graph, execute wave-by-wave |
+| `tasks.json`    | Proceed to PRD task loop below                             |
+| Only `spec.md`  | Convert spec to `tasks.json`, then proceed                    |
 
 ## Phase 3: Wave-Based Execution
 
 If `plan.md` exists with dependency graph:
 
-1. **Load skill:** `skill({ name: "executing-plans" })`
-2. **Parse waves** from dependency graph section
+1. **Parse waves** from dependency graph section
 3. **Execute wave-by-wave:**
    - Single-task wave → execute directly (no subagent overhead)
    - Multi-task wave → dispatch parallel `task({ subagent_type: "general" })` subagents, one per task
@@ -122,8 +87,8 @@ For each task (wave-based or sequential fallback):
 6. **Verify** — run each verification step from the task
 7. **If verification fails**, fix and retry (max 2 attempts per task)
 8. **Commit** — per-task commit (see below)
-9. **Mark** `passes: true` in `prd.json`
-10. **Append** progress to `.beads/artifacts/$ARGUMENTS/progress.txt`
+9. **Mark** `passes: true` in `.opencode/artifacts/tasks.json`
+10. **Append** progress to `.opencode/artifacts/$(cat .opencode/artifacts/.active)/progress.md`
 
 ### Checkpoint Protocol
 
@@ -196,7 +161,7 @@ After each task completes (verification passed):
 3. **Commit with type prefix:**
 
    ```bash
-   git commit -m "feat(bead-$ARGUMENTS): [task description]
+   git commit -m "feat: [task description]
 
    - [key change 1]
    - [key change 2]"
@@ -228,7 +193,7 @@ Follow the [Verification Protocol](../skill/verification-before-completion/refer
 - All 4 gates must pass before proceeding to commit/push
 - Also run PRD `Verify:` commands
 
-If the PRD requires local web, browser, OAuth callback, webhook, or multi-service verification, load the [portless](../skill/portless/SKILL.md) skill and use approved stable URLs as verification evidence. Portless is optional: read-only `portless list` / `portless get <service>` checks are allowed when installed, but do not install Portless, start proxies, trust CAs, mutate hosts files, clean Portless state, or expose LAN services without explicit user approval.
+If the PRD requires local web, browser, OAuth callback, webhook, or multi-service verification, use stable URLs as verification evidence.
 
 ## Phase 5: Review
 
@@ -237,9 +202,20 @@ BASE_SHA=$(git rev-parse origin/main 2>/dev/null || git rev-parse HEAD~1)
 HEAD_SHA=$(git rev-parse HEAD)
 ```
 
-### UI Quality Gate (if UI files changed)
+### Mode Selection
+
+| Condition | Mode |
+|---|---|
+| Routine change, low risk | Standard Review (below) |
+| High-risk feature, explicit user request for quality gating, or the build agent flagged the feature as requiring iterative quality gating | Iterative Quality Loop |
 
-Before general review, detect changed UI files:
+When using Standard Review mode, apply the UI Quality Gate then the parallel review. When using Iterative Loop mode, apply the UI Quality Gate first (before entering the loop), then run the scored loop flow.
+
+---
+
+### UI Quality Gate (always — both modes)
+
+Detect changed UI files:
 
 ```bash
 git diff --name-only $BASE_SHA...HEAD -- \
@@ -248,29 +224,26 @@ git diff --name-only $BASE_SHA...HEAD -- \
 
 If any UI files changed:
 
-1. Load `skill({ name: "ux-quality-gates" })`.
-2. Run `/ui-slop-check auto --since=$BASE_SHA` or manually apply its checklist when slash-command invocation is unavailable.
-3. Verify UX gates for changed surfaces:
+1. Run `/ui-slop-check auto --since=$BASE_SHA` or manually apply its checklist when slash-command invocation is unavailable.
+2. Verify UX gates for changed surfaces:
    - One primary action per view/section
    - Empty/loading/error/success states for async/data flows
    - Retry/undo/confirm paths for errors and destructive actions
    - Form labels, helper text, validation, and error association
    - Semantic HTML, keyboard path, visible focus, reduced motion
    - Component family consistency for related controls
-4. Treat Critical findings like review Critical findings: fix inline, rerun verification, then continue.
+3. Treat Critical findings like review Critical findings: fix inline, rerun verification, then continue.
 
-Load and run the review skill:
+---
 
-```typescript
-skill({ name: "requesting-code-review" });
-```
+### Standard Review Mode
 
-Run **5 parallel agents**: security/correctness, performance/architecture, type-safety/tests, conventions/patterns, simplicity/completeness.
+Run **5 parallel agents** for review: security/correctness, performance/architecture, type-safety/tests, conventions/patterns, simplicity/completeness.
 
 Fill placeholders:
 
-- `{WHAT_WAS_IMPLEMENTED}`: bead title + brief summary of what changed
-- `{PLAN_OR_REQUIREMENTS}`: `.beads/artifacts/$ARGUMENTS/prd.md`
+- `{WHAT_WAS_IMPLEMENTED}`: brief summary of what changed
+- `{PLAN_OR_REQUIREMENTS}`: `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md`
 - `{BASE_SHA}` / `{HEAD_SHA}`: from above
 
 Wait for all 5 agents to return. Synthesize findings.
@@ -279,10 +252,98 @@ Wait for all 5 agents to return. Synthesize findings.
 
 - Critical issues → fix inline, re-run Phase 4 verification, continue
 - Important issues → fix inline, continue
-- Minor issues → add to bead comments, note for `/compound` step
+- Minor issues → note in `.opencode/artifacts/$(cat .opencode/artifacts/.active)/progress.md`, flag for `/compound` step
 
 If review finds critical issues that require architectural decisions → stop → present options to user.
 
+### Iterative Quality Loop Mode
+
+Score-gated feedback loop for high-risk features. Replaces the standard parallel review with a structured iteration cycle.
+
+#### Setup
+
+Initialize loop state:
+
+```bash
+SLUG=$(cat .opencode/artifacts/.active)
+cat > ".opencode/artifacts/$SLUG/review-state.json" << EOF
+{
+  "slug": "$SLUG",
+  "rounds": 0,
+  "maxRounds": 5,
+  "lastScore": 0,
+  "sameScoreCount": 0,
+  "findingsResolved": 0,
+  "findingsRemaining": 0,
+  "status": "active"
+}
+EOF
+```
+
+#### Loop
+
+Repeat steps 2-8 until exit or escalation:
+
+| Step | Action |
+|---|---|
+| **1. EXECUTE** | Implement per spec/plan (already done in Phase 3) |
+| **2. REVIEW** | Spawn **one review subagent** (`subagent_type: "review"`) with spec + current diff + `review-state.json`. Returns: score (X/5), findings array (severity + file:line + suggestion), suggested next action |
+| **3. GATE** | Score ≥ 5 → mark done (`status: "passed"`), exit loop, proceed to Goal-Backward Verification. Score 4 → ask user if they want to proceed or loop. Score <4 → continue |
+| **4. STALL?** | If `sameScoreCount ≥ 2` → escalate: surface accumulated findings, present to user with a recommendation |
+| **5. MAX?** | If `rounds ≥ maxRounds` → escalate with full finding log |
+| **6. FILTER** | Split findings into categories and handle each: |
+| | • **Actionable** (code-level, clear fix) → proceed to fix |
+| | • **Informational** (note, no code change) → log to progress.md with `[info]` |
+| | • **Architecture/Design** → stop loop, present to user for decision |
+| **7. FIX** | For each actionable finding, spawn a fix subagent with the exact file:line and suggested fix. Run sequentially for same-file findings, parallel for different files |
+| **8. RE-REVIEW** | Update `review-state.json`: increment rounds, update score, reset/resolve findings. Go to step 2 |
+
+#### Loop State Updates
+
+After each round, update `review-state.json`:
+
+**`sameScoreCount` rule:**
+- If new score === `lastScore` → increment `sameScoreCount`
+- If new score !== `lastScore` → reset `sameScoreCount` to 0
+
+**Example after round 1 (score: 3):**
+
+```json
+{
+  "rounds": 1,
+  "lastScore": 3,
+  "sameScoreCount": 0,
+  "findingsResolved": 2,
+  "findingsRemaining": 1,
+  "status": "active"
+}
+```
+
+**Status transitions:**
+
+- Stall detected (`sameScoreCount ≥ 2`) → `status: "stalled"`, append accumulated findings to progress.md
+- Max rounds reached → `status: "maxed"`, append full finding log to progress.md
+- Pass (score ≥ 5) → `status: "passed"`, proceed to Goal-Backward Verification
+
+#### Review Subagent Prompt
+
+When spawning, include:
+
+- The original spec/slug
+- The current diff (all changed files since the start of Phase 3)
+- The current `review-state.json`
+- Return format: `{ score: number, findings: Array<{severity:"critical"|"important"|"minor", file:string, line:number, suggestion:string, type:"actionable"|"informational"|"architecture"}>, nextAction: string }`
+
+#### Exit Conditions
+
+| Condition | Action |
+|---|---|
+| Score ≥ 5 | Proceed to Goal-Backward Verification |
+| User approves score 4 | Proceed to Goal-Backward Verification |
+| Architecture finding | Stop, present options to user |
+| Stalled (same score 2x) | Escalate with accumulated findings |
+| Max rounds | Escalate with full finding log |
+
 ### Goal-Backward Verification (if plan.md exists)
 
 Verify that tasks completed ≠ goals achieved:
@@ -325,10 +386,10 @@ question({
   questions: [
     {
       header: "Close",
-      question: "All tasks pass, gates green, review clean. Close bead $ARGUMENTS?",
+      question: "All tasks pass, gates green, review clean. Mark plan as complete?",
       options: [
-        { label: "Yes, close it (Recommended)", description: "All checks passed" },
-        { label: "No, keep open", description: "Need more work" },
+        { label: "Yes, mark complete (Recommended)", description: "All checks passed" },
+        { label: "No, keep working", description: "Need more work" },
       ],
     },
   ],
@@ -337,12 +398,9 @@ question({
 
 If confirmed:
 
-```bash
-br close $ARGUMENTS --reason "Shipped: all PRD tasks pass, verification + review passed"
-br sync --flush-only
-```
+Update `.opencode/artifacts/todo.md` to mark all tasks complete and append summary to `.opencode/artifacts/$(cat .opencode/artifacts/.active)/progress.md`.
 
-Record significant learnings with `/compound $ARGUMENTS` after closing.
+Record significant learnings with `/compound` after closing.
 
 ## Output
 
@@ -380,7 +438,7 @@ Report:
 6. **Next Steps:**
    - `/pr` to create pull request
    - Manual commits if not already done
-   - Create follow-up beads for deferred work
+   - Note deferred work in `.opencode/artifacts/todo.md`
 
 ## Related Commands
 

package/dist/template/.opencode/command/ui-review.md

@@ -10,10 +10,8 @@ model: proxypal/gemini-3-pro-preview
 ## Load Skills
 
 ```typescript
-skill({ name: "visual-analysis" }); // Analysis framework
 skill({ name: "accessibility-audit" }); // WCAG checklists
 skill({ name: "frontend-design" }); // Anti-patterns, design quality
-skill({ name: "ux-quality-gates" }); // IA, heuristics, forms, recovery, state coverage
 ```
 
 ## Input
@@ -28,7 +26,7 @@ Parse `$ARGUMENTS`:
 
 ### 1. Analyze the Input
 
-Use the `visual-analysis` skill to perform deep analysis:
+Analyze the input:
 
 - Content inventory (elements, text, icons)
 - Visual properties (colors, typography, spacing, layout)
@@ -36,7 +34,7 @@ Use the `visual-analysis` skill to perform deep analysis:
 
 ### 2. Usability Heuristic Pass
 
-Apply `ux-quality-gates` before visual scoring. Check:
+Before visual scoring, check:
 
 - System status is visible for async or multi-step work
 - Labels and navigation use user vocabulary, not implementation terms

package/dist/template/.opencode/command/verify.md

@@ -1,6 +1,6 @@
 ---
 description: Verify implementation completeness, correctness, and coherence
-argument-hint: "<bead-id> [--quick] [--full] [--fix] [--no-cache]"
+argument-hint: "[path|all] [--quick] [--full] [--fix] [--no-cache]"
 agent: review
 ---
 
@@ -11,16 +11,14 @@ Check implementation against PRD before shipping.
 ## Load Skills
 
 ```typescript
-skill({ name: "beads" });
 skill({ name: "verification-before-completion" });
-// If local web/browser verification needs stable URLs: skill({ name: "portless" });
 ```
 
 ## Parse Arguments
 
 | Argument     | Default  | Description                                    |
 | ------------ | -------- | ---------------------------------------------- |
-| `<bead-id>`  | required | The bead to verify                             |
+| `<path\|all>`| required | The path or keyword to verify                  |
 | `--quick`    | false    | Gates only, skip coherence check               |
 | `--full`     | false    | Force full verification mode (non-incremental) |
 | `--fix`      | false    | Auto-fix lint/format issues                    |
@@ -28,11 +26,10 @@ skill({ name: "verification-before-completion" });
 
 ## Determine Input Type
 
-| Input Type | Detection                   | Action                              |
-| ---------- | --------------------------- | ----------------------------------- |
-| Bead ID    | Matches `br-xxx` or numeric | Check implementation vs PRD in bead |
-| Path       | File/directory path         | Verify that specific path           |
-| `all`      | Keyword                     | Verify all in-progress work         |
+| Input Type | Detection           | Action                     |
+| ---------- | ------------------- | -------------------------- |
+| Path       | File/directory path | Verify that specific path  |
+| `all`      | Keyword             | Verify all in-progress work |
 
 ## Before You Verify
 
@@ -51,7 +48,7 @@ CURRENT_STAMP=$(printf '%s\n%s' \
   "$(git rev-parse HEAD)" \
   "$(git diff HEAD -- '*.ts' '*.tsx' '*.js' '*.jsx')" \
   | shasum -a 256 | cut -d' ' -f1)
-LAST_STAMP=$(tail -1 .beads/verify.log 2>/dev/null | awk '{print $1}')
+LAST_STAMP=$(tail -1 .opencode/artifacts/verify.log 2>/dev/null | awk '{print $1}')
 ```
 
 | Condition                                 | Action                                                 |
@@ -68,19 +65,16 @@ Verification: cached PASS (no changes since <timestamp from verify.log>)
 
 ## Phase 1: Gather Context
 
-```bash
-br show $ARGUMENTS
-```
+Read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/spec.md` to understand the requirements.
 
-Read `.beads/artifacts/$ARGUMENTS/` to check what artifacts exist.
+Read `.opencode/artifacts/$(cat .opencode/artifacts/.active)/` to check what plan artifacts exist.
 
 Read the PRD and any other artifacts (plan.md, research.md, design.md).
 
 **Verify guards:**
 
-- [ ] Bead is `in_progress`
-- [ ] `prd.md` exists
-- [ ] You have read the full PRD
+- [ ] Plan/spec exists and is up to date
+- [ ] You have read the full spec
 
 ## Phase 2: Completeness
 
@@ -106,7 +100,7 @@ Follow the [Verification Protocol](../skill/verification-before-completion/refer
 1. **Parallel**: typecheck + lint (simultaneously)
 2. **Sequential** (after parallel passes): test, then build (ship only)
 
-For browser/manual local-web requirements, load the [portless](../skill/portless/SKILL.md) skill when stable named URLs would improve verification. Use only read-only Portless commands unless the user explicitly approves local networking changes; a reachable Portless URL supplements, but never replaces, typecheck/lint/test/build evidence.
+For browser/manual local-web requirements, use stable URLs as verification evidence. A reachable URL supplements, but never replaces, typecheck/lint/test/build evidence.
 
 Report results with mode column:
 
@@ -122,7 +116,7 @@ Report results with mode column:
 **After all gates pass**, record to verification cache:
 
 ```bash
-echo "$CURRENT_STAMP $(date -u +%Y-%m-%dT%H:%M:%SZ) PASS" >> .beads/verify.log
+echo "$CURRENT_STAMP $(date -u +%Y-%m-%dT%H:%M:%SZ) PASS" >> .opencode/artifacts/verify.log
 ```
 
 If `--fix` flag provided, run the project's auto-fix command (e.g., `npm run lint:fix`, `ruff check --fix`, `cargo clippy --fix`).
@@ -139,9 +133,7 @@ Flag contradictions with specific file references.
 
 ## Phase 5: Report
 
-```bash
-br comments add $ARGUMENTS "Verification: [PASS|PARTIAL|FAIL] - [summary]"
-```
+Append to `.opencode/artifacts/$(cat .opencode/artifacts/.active)/progress.md`: `Verification: [PASS|PARTIAL|FAIL] - [summary]`
 
 Output:
 
@@ -157,7 +149,7 @@ Record significant findings with `observation()`:
 ```typescript
 observation({
   type: "discovery", // or "warning", "bugfix"
-  title: "Verify: [bead-id] [key finding]",
+  title: "Verify: [scope] [key finding]",
   narrative: "[What was found, impact, resolution]",
   concepts: "verification, [component]",
   confidence: "high",