[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78GSQGKBBJNQRT3DF10QG9T":3,"skill-same:01M78GSQGKBBJNQRT3DF10QG9T":52},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M78GSQGKBBJNQRT3DF10QG9T","review-work",null,"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal\u002Fconstraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub\u002Fgit\u002FSlack\u002FNotion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","younghai\u002Flazyz","younghai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F10753869?v=4","01M78GJ5KADQ959GDR0ZM7DDNF","young kwon","01M78GJ5M3F6D2YKVXCNX3WMR9",0,"MIT","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Frelease\u002Flazyz-0.11.0-20261007\u002Fskills\u002Freview-work\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: review-work\ndescription: \"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal\u002Fconstraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub\u002Fgit\u002FSlack\u002FNotion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.\"\n---\n\n## ZCode Harness Tool Compatibility\n\nThis skill was ported from the Codex\u002FOpenCode OmO harness. ZCode does not provide `call_omo_agent`, `task`, `background_output`, `team_*`, `multi_agent_v1.spawn_agent`, or any Codex-only multi-agent tools. Translate every such example to ZCode's native tools:\n\n| Codex\u002FOpenCode example | ZCode tool to use |\n| --- | --- |\n| `call_omo_agent(subagent_type=\"explore\", ...)` | `Agent` tool with `subagent_type=\"Explore\"` and a self-contained `prompt` |\n| `call_omo_agent(subagent_type=\"librarian\", ...)` | `Agent` tool with `subagent_type=\"Explore\"` (research-focused prompt) |\n| `task(subagent_type=\"plan\", ...)` | `Agent` tool with `subagent_type=\"general-purpose\"` (planning-focused prompt) |\n| `task(subagent_type=\"oracle\", ...)` for final verification | `Agent` tool with `subagent_type=\"general-purpose\"` (rigorous-review prompt) |\n| `task(category=\"...\", ...)` for implementation or QA | `Agent` tool with `subagent_type=\"general-purpose\"` |\n| `multi_agent_v1.spawn_agent(...)` \u002F `spawn_agent(...)` | `Agent` tool with `subagent_type` and a self-contained `prompt` |\n| `background_output(task_id=\"...\")` | Only unstructured read-only research may use background mode; its completion arrives as a parent task notification, not a plugin hook |\n| `multi_agent_v1.wait_agent(...)` mailbox signals | No equivalent plugin tool; use returned foreground Agent results or the host's background research notifications |\n| `team_*(...)` | ZCode has no native team-of-threads API. Approximate with multiple parallel `Agent` calls (one per member) coordinated by the parent, or fall back to sequential subagents when isolation is unclear |\n\nZCode v3.14.3 loads enabled plugin agents from `agents\u002F`. Prefer namespaced roles such as `lazyz:explorer` and `lazyz:lazycodex-executor`; unambiguous bare names also work. Include all task context and ownership in the self-contained `prompt`. Roles inherit the user's active model; do not force an obsolete provider ID or copy Codex model\u002Feffort fields. If an older host lacks a role, describe it inline using the appropriate built-in agent.\n\nBefore every governed Agent or Task call, read the Dispatch schema in the installed [start-work skill](..\u002Fstart-work\u002FSKILL.md) and prepend exactly one LAZYCODEX_DISPATCH_V1 JSON line. General-purpose and known executor calls require this envelope; read-only Explore calls may remain unstructured. Use the same taskId across retries with a fresh runId, unchanged criteria, role, phase and budget. Planning\u002Fdocument work uses writer, code work implementer, review reviewer, and testing qa. Never rename a task to bypass its exhausted budget.\n\nGoverned calls must use `run_in_background: false` (or omit it). A background launch is not a completed result and has no matching plugin completion hook. Multiple independent foreground calls may be requested together; do not assume the host executes them concurrently. Read-only background research without a dispatch contract remains available.\n\nRequire each governed child to write the matching ReceiptV1 with current source identity, criterion invocations, zero exit codes and hashed regular artifacts, then return EVIDENCE_RECORDED; final-phase work requires FINAL_EVIDENCE_RECORDED. PostToolUse records failed evidence; Stop prevents unresolved completion while budget remains. An exhausted budget requires a blocker handoff, not an unsupported completion claim.\n\nUser authorization and verification policy take precedence over generic workflow instructions below. Do not auto-commit, push, merge, delete, deploy, install tools, call paid services, or run checks prohibited by the user. Record skipped checks honestly; do not fabricate a passing receipt. A task requiring a verified receipt remains unverified if the necessary checks are prohibited. ZCode Stop can continue at most three consecutive times, not indefinitely. Native Goal Mode, Memory and Wiki are host features; no callable API is assumed.\n\nFor work likely to take long, prefer multiple concurrent `Agent` calls in a single message over one giant blocking call. Treat each child's final message as the deliverable: if it lacks the required output, ack-only, or stops without producing evidence, fall back to a smaller re-scoped `Agent` call with the missing deliverable made explicit in the `prompt`.\n\n## Codex Subagent Reliability\n\nEvery `spawn_agent` message must be self-contained. Start with\n`TASK: \u003Cimperative assignment>`, then name `DELIVERABLE`, `SCOPE`, and\n`VERIFY`. State that it is an executable assignment, not a context\nhandoff. Role selection requires `agent_type`; `model` +\n`reasoning_effort` alone creates a default agent, not a reviewer or\nworker. Prefer `fork_turns: \"none\"` unless full history is truly\nrequired; paste only the review context that worker needs.\n\nPlan and reviewer agents may run for a long time; spawn them in the background, keep doing independent root work, and poll with short wait_agent cycles sized to the work. Never use a single long blocking wait for them, and never spin on tiny timeouts as a failure budget.\n\nTreat child status as a progress signal, not a timeout counter. For\nwork likely to exceed one wait cycle, require the child to send\n`WORKING: \u003Ctask> - \u003Ccurrent phase>` before long reading, testing, or\nreview passes, and `BLOCKED: \u003Creason>` only when it cannot progress.\nWhile any child is active, keep the parent visibly alive with active\nsubagent count, agent names, latest `WORKING:` phase, and whether the\nparent is waiting for mailbox updates. Track spawned agent names\nlocally. Use `wait_agent` for mailbox signals, not proof of completion.\nA timeout only means no new mailbox update arrived; after a timeout,\nrun a single `list_agents` check for the named child when you need\nreassurance. If it is running or its latest message is `WORKING:`,\ntreat it as alive. Do not use `list_agents` as a polling loop or status\nfeed; it can replay large payloads. Fallback only when the child is\ncompleted without the deliverable, ack-only after followup, explicitly\n`BLOCKED:`, or no longer running. Then mark that review lane\ninconclusive, do not count it as PASS or approval, close if safe, and\nrespawn a smaller `fork_turns: \"none\"` reviewer with the missing\ndeliverable.\n\n# Review Work - 5-Agent Parallel Review Orchestrator\n\nLaunch 5 specialized sub-agents in parallel to review completed implementation work from every angle. All 5 must pass for the review to pass. If even ONE fails, the review fails.\n\nWhen `review-work` is used as a final implementation, PR, or `$start-work`\ngate, it is blocking. A timeout, missing deliverable, ack-only response,\nexplicit `BLOCKED:`, or inconclusive lane is not a pass. Treat that lane as\nfailed, investigate the underlying uncertainty with the `debugging` skill when\nruntime behavior may be wrong, fix with evidence, and rerun the affected lane\nbefore claiming completion, creating or handing off a PR, or merging.\n\nA rejecting lane must name its blockers inline in its final message — each\nblocker cites the violated goal criterion or requirement plus an evidence\npointer. A bare REJECT\u002FFAIL token without findings is not a verdict; treat it\nas an inconclusive lane (one bounded respawn, then record it inconclusive with\nthat reason).\n\nWhen reviewing a PR or branch, collect diff, file contents, and verification\nresults from a dedicated review worktree attached to that branch. Never\ncheckout, test, or edit the review branch in the main worktree.\n\nReview evidence must be safe to share. Redact or mask secrets and sensitive\nuser data before including evidence in logs, PR bodies, or handoffs. Never\ninclude raw tokens, credentials, auth headers, cookies, API keys, env dumps,\nprivate logs, or PII; summarize with lengths, hashes, and short non-sensitive\nprefixes when identity is needed.\n\nThe 5 agents cover complementary concerns - together they form a comprehensive review that no single reviewer could match:\n\n| # | Agent | Type | Role | Focus Level |\n|---|-------|------|------|-------------|\n| 1 | Goal Verifier | Oracle | Did we build what was asked? | MAIN |\n| 2 | QA Executor | unspecified-high | Does it actually work? | MAIN |\n| 3 | Code Reviewer | Oracle | Is the code well-written? | MAIN |\n| 4 | Security Auditor | Oracle | Is it secure? | SUB |\n| 5 | Context Miner | unspecified-high | Did we miss any context? | MAIN |\n\n---\n\n## Phase 0: Gather Review Context\n\nBefore launching agents, collect these inputs. Extract from conversation history first - the user's original request, constraints discussed, and decisions made are usually already in the thread. Only ask if truly missing.\n\n\u003Crequired_inputs>\n\n- **GOAL**: The original objective. What was the user trying to achieve? Pull from the initial request in this conversation.\n- **CONSTRAINTS**: Rules, requirements, or limitations. Tech stack restrictions, performance targets, API contracts, design patterns to follow, backward compatibility needs.\n- **BACKGROUND**: Why this work was needed. Business context, user stories, related systems, prior decisions that informed the approach.\n- **CHANGED_FILES**: Auto-collect via `git diff --name-only HEAD~1` or against the appropriate base (branch point, specific commit).\n- **DIFF**: Auto-collect via `git diff HEAD~1` or against the appropriate base.\n- **FILE_CONTENTS**: Read the full content of each changed file (not just the diff). Oracle agents cannot read files - they need full context in the prompt.\n- **RUN_COMMAND**: How to start\u002Frun the application. Check `package.json` scripts, `Makefile`, `docker-compose.yml`, or ask the user.\n\n\u003C\u002Frequired_inputs>\n\n\n**NEVER CHECKOUT A PR BRANCH IN THE MAIN WORKTREE. ALWAYS CREATE A NEW GIT WORKTREE (`git worktree add`) AND WORK THERE. THIS PREVENTS CONTAMINATING THE USER'S WORKING DIRECTORY WITH UNRELATED BRANCH STATE.**\n\n**Auto-collection sequence:**\n\n```bash\n# 1. Get changed files\ngit diff --name-only HEAD~1  # or: git diff --name-only main...HEAD\n\n# 2. Get diff\ngit diff HEAD~1  # or: git diff main...HEAD\n\n# 3. Detect run command\n# Check package.json -> \"scripts.dev\" or \"scripts.start\"\n# Check Makefile -> default target\n# Check docker-compose.yml -> services\n```\n\nFor GOAL, CONSTRAINTS, BACKGROUND - review the full conversation history. The user's original message almost always contains the goal. Constraints often emerge during discussion. If anything critical is ambiguous, ask ONE focused question - not a checklist.\n\n---\n\n## Phase 1: Launch 5 Agents\n\nLaunch ALL 5 in a single turn. Every agent uses `run_in_background=true`. No sequential launches. No waiting between them.\n\n**Oracle agents receive everything in the prompt** (they cannot read files or run commands). Include DIFF + FILE_CONTENTS + all context directly in the prompt text.\n\n**unspecified-high agents are autonomous** - they can read files, run commands, and use tools. Give them goals and pointers, not raw content dumps.\n\n---\n\n### Agent 1: Goal & Constraint Verification (Oracle) - MAIN\n\nThis agent answers: \"Did we build exactly what was asked, within the rules we were given?\"\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Verify implementation against original goal and constraints\",\n  prompt=\"\"\"\n\u003Creview_type>GOAL & CONSTRAINT VERIFICATION\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL - paste the user's original request and any clarifications}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS - every rule, requirement, or limitation discussed}\n\u003C\u002Fconstraints>\n\n\u003Cbackground>\n{BACKGROUND - why this work was needed, broader context}\n\u003C\u002Fbackground>\n\n\u003Cchanged_files>\n{CHANGED_FILES - list of modified file paths}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of every changed file, clearly delimited per file}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF - the actual git diff}\n\u003C\u002Fdiff>\n\nReview whether this implementation correctly and completely achieves the stated goal within the given constraints. Be obsessively thorough - the point of this review is to catch what the implementer missed.\n\nREVIEW CHECKLIST:\n\n1. **Goal Completeness**: Break the goal into every sub-requirement (explicit AND implied). For each, mark ACHIEVED \u002F MISSED \u002F PARTIAL. Missing even one implied requirement that a reasonable engineer would have addressed = PARTIAL at minimum.\n\n2. **Constraint Compliance**: List every constraint. For each, verify compliance with specific code evidence. A constraint violated = automatic FAIL.\n\n3. **Requirement Gaps**: Requirements the user clearly wanted but didn't spell out. Things implied by the goal or background that a thoughtful engineer would have included.\n\n4. **Over-Engineering**: Anything added that wasn't requested - unnecessary abstractions, extra features, premature optimizations, speculative generality. Flag these as scope creep.\n\n5. **Edge Cases**: Given the goal, what inputs or scenarios would break this? Trace through at least 5 edge cases mentally.\n\n6. **Behavioral Correctness**: Walk through the code logic for 3+ representative scenarios. Does the code actually produce the expected behavior in each case?\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cgoal_breakdown>\n  For each sub-requirement:\n  - [ACHIEVED\u002FMISSED\u002FPARTIAL] Requirement description\n  - Evidence: specific code reference or gap\n\u003C\u002Fgoal_breakdown>\n\u003Cconstraint_compliance>\n  For each constraint:\n  - [ACHIEVED\u002FMISSED] Constraint description - evidence\n\u003C\u002Fconstraint_compliance>\n\u003Cfindings>\n  - [PASS\u002FFAIL\u002FWARN] Category: Description\n  - File: path (line range if applicable)\n  - Evidence: specific code or logic reference\n\u003C\u002Ffindings>\n\u003Cblocking_issues>Issues that MUST be fixed. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 2: QA via App Execution (unspecified-high) - MAIN\n\nThis agent answers: \"Does it actually work when you run it?\"\n\nThe QA agent follows a structured process: brainstorm scenarios exhaustively first, then self-review and augment, then create a task list, then execute systematically.\n\n```\ntask(\n  category=\"unspecified-high\",\n  run_in_background=true,\n  load_skills=[\"playwright\", \"dev-browser\"],\n  description=\"QA by actually running and using the application\",\n  prompt=\"\"\"\n\u003Creview_type>QA - HANDS-ON APP EXECUTION\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS}\n\u003C\u002Fconstraints>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Crun_command>\n{RUN_COMMAND - how to start the application, or \"unknown\" if not determined}\n\u003C\u002Frun_command>\n\nYou are a QA engineer. Your job is to RUN the application and verify it works through hands-on testing. You do not review code - you test behavior.\n\nMANDATORY PROCESS (follow in order):\n\n### Step 1: Scenario Brainstorm\n\nBefore touching the app, write down EVERY test scenario you can think of. Be exhaustive. Think about:\n\n- **Happy paths**: The primary use cases this implementation enables. What's the main thing the user wanted to do?\n- **Boundary conditions**: Empty inputs, maximum-length inputs, zero values, negative numbers, special characters, unicode, very large datasets.\n- **Error paths**: Invalid inputs, network failures, missing files, permission denied, timeout conditions.\n- **Regression scenarios**: Existing features that touch the same code paths. Things that worked before and must still work.\n- **State transitions**: What happens when you do things out of order? Rapid repeated actions? Concurrent usage?\n- **UX scenarios** (if applicable): Layout on different sizes, keyboard navigation, screen reader compatibility, loading states, error messages.\n- **Integration points**: Does this feature interact with external services, databases, or other modules? Test those boundaries.\n\nWrite each scenario as a one-liner with expected behavior. Aim for 15-30 scenarios minimum.\n\n### Step 2: Scenario Augmentation\n\nReview your scenario list with fresh eyes. For each scenario, ask:\n- \"What could go wrong here that I haven't considered?\"\n- \"What would a malicious or careless user do?\"\n- \"What environmental conditions could affect this?\" (disk full, slow network, expired tokens)\n\nAdd at least 5 more scenarios from this reflection. Group scenarios by priority: P0 (must pass), P1 (should pass), P2 (nice to pass).\n\n### Step 3: Create Task List\n\nConvert your augmented scenario list into a structured task list (use TaskCreate\u002FTaskUpdate or your todo system). Each task = one test scenario with:\n- Test name\n- Steps to execute\n- Expected result\n- Priority (P0\u002FP1\u002FP2)\n\n### Step 4: Execute Systematically\n\nWork through the task list in priority order (P0 first). For each test:\n\n1. Execute the test steps\n2. Record actual result\n3. Compare with expected result\n4. Mark PASS or FAIL\n5. If FAIL: capture evidence (screenshot, terminal output, error message)\n6. Mark the task complete\n\n**Execution guidance by app type:**\n- **Web app**: Use playwright\u002Fdev-browser to navigate, click, fill forms, verify visual output.\n- **CLI tool**: Run commands with various arguments, pipe inputs, check exit codes and output.\n- **Library\u002FSDK**: Write and execute a test script that imports and exercises the public API.\n- **Backend API**: Use curl\u002Fhttpie to hit endpoints with various payloads, verify response codes and bodies.\n- **Mobile\u002FDesktop**: If not directly runnable, write integration tests and execute them.\n\nIf the app cannot be started (build failure), that's an immediate FAIL - no need to continue.\n\n### Step 5: Compile Results\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cscenario_coverage>\n  Total scenarios: N\n  P0: X tested, Y passed\n  P1: X tested, Y passed\n  P2: X tested, Y passed\n\u003C\u002Fscenario_coverage>\n\u003Ctest_results>\n  For each test:\n  - [PASS\u002FFAIL] Test name (Priority)\n  - Steps: What you did\n  - Expected: What should happen\n  - Actual: What actually happened\n  - Evidence: Screenshot path or terminal output snippet (if FAIL)\n\u003C\u002Ftest_results>\n\u003Cblocking_issues>P0 or P1 failures only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 3: Code Quality Review (Oracle) - MAIN\n\nThis agent answers: \"Is the code well-written, maintainable, and consistent with the codebase?\"\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Review overall code quality, patterns, and architecture\",\n  prompt=\"\"\"\n\u003Creview_type>CODE QUALITY REVIEW\u003C\u002Freview_type>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of changed files AND neighboring files that show existing patterns}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF}\n\u003C\u002Fdiff>\n\n\u003Cbackground>\n{BACKGROUND}\n\u003C\u002Fbackground>\n\nYou are a senior staff engineer conducting a code review. Your standard: \"Would I approve this PR without comments?\"\n\nREVIEW DIMENSIONS (examine each):\n\n1. **Correctness**: Logic errors, off-by-one, null\u002Fundefined handling, race conditions, resource leaks, unhandled promise rejections.\n\n2. **Pattern Consistency**: Does new code follow the codebase's established patterns? Compare with the neighboring files provided. Introducing a new pattern where one already exists = finding.\n\n3. **Naming & Readability**: Clear variable\u002Ffunction\u002Ftype names? Self-documenting code? Would another engineer understand this without explanation?\n\n4. **Error Handling**: Errors properly caught, logged, and propagated? No empty catch blocks? No swallowed errors? User-facing errors are helpful?\n\n5. **Type Safety**: Any `as any`, `@ts-ignore`, `@ts-expect-error`? Proper generic usage? Correct type narrowing? (If TypeScript\u002Ftyped language)\n\n6. **Performance**: N+1 queries? Unnecessary re-renders? Blocking I\u002FO on hot paths? Memory leaks? Unbounded growth?\n\n7. **Abstraction Level**: Right level of abstraction? No copy-paste duplication? But also no premature over-abstraction?\n\n8. **Testing**: New behaviors covered by tests? Tests are meaningful, not just coverage padding? Test names describe scenarios?\n\n9. **API Design**: Public interfaces clean and consistent with existing APIs? Breaking changes flagged?\n\n10. **Tech Debt**: Does this introduce new tech debt? Or create coupling that will be painful to change?\n\nCategorize each finding by severity:\n- **CRITICAL**: Will cause bugs, data loss, or crashes in production\n- **MAJOR**: Significant quality issue that should be fixed before merge\n- **MINOR**: Improvement worth making but not blocking\n- **NITPICK**: Style preference, optional\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cfindings>\n  - [CRITICAL\u002FMAJOR\u002FMINOR\u002FNITPICK] Category: Description\n  - File: path (line range)\n  - Current: what the code does now\n  - Suggestion: how to improve\n\u003C\u002Ffindings>\n\u003Cblocking_issues>CRITICAL and MAJOR items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 4: Security Review (Oracle) - SUB\n\nThis agent answers: \"Are there security vulnerabilities in these changes?\"\n\nThis is supplementary - it focuses exclusively on security. It does NOT comment on code style, architecture, or functionality unless those directly create a security risk.\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Security-focused review of implementation changes\",\n  prompt=\"\"\"\n\u003Creview_type>SECURITY REVIEW (supplementary)\u003C\u002Freview_type>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of changed files}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF}\n\u003C\u002Fdiff>\n\nYou are a security engineer. Review this diff exclusively for security vulnerabilities and anti-patterns. Ignore code style, naming, architecture - unless it directly creates a security risk.\n\nSECURITY CHECKLIST:\n\n1. **Input Validation**: User inputs sanitized? SQL injection, XSS, command injection, SSRF vectors?\n2. **Auth & AuthZ**: Authentication checks where needed? Authorization verified for each action? Privilege escalation paths?\n3. **Secrets & Credentials**: Hardcoded secrets, API keys, tokens in code or config? Secrets in logs?\n4. **Data Exposure**: Sensitive data in logs? PII in error messages? Over-exposed API responses?\n5. **Dependencies**: New dependencies added? Known CVEs? Suspicious or unnecessary packages?\n6. **Cryptography**: Proper algorithms? No custom crypto? Secure random? Proper key management?\n7. **File & Path**: Path traversal? Unsafe file operations? Symlink following?\n8. **Network**: CORS configured correctly? Rate limiting? TLS enforced? Certificate validation?\n9. **Error Leakage**: Stack traces exposed to users? Internal details in error responses?\n10. **Supply Chain**: Lockfile updated consistently? Dependency pinning?\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cseverity>CRITICAL \u002F HIGH \u002F MEDIUM \u002F LOW \u002F NONE\u003C\u002Fseverity>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cfindings>\n  - [CRITICAL\u002FHIGH\u002FMEDIUM\u002FLOW] Category: Description\n  - File: path (line range)\n  - Risk: What could an attacker do?\n  - Remediation: Specific fix\n\u003C\u002Ffindings>\n\u003Cblocking_issues>CRITICAL and HIGH items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 5: Context Mining (unspecified-high) - MAIN\n\nThis agent answers: \"Did we miss any context that should have informed this implementation?\"\n\n```\ntask(\n  category=\"unspecified-high\",\n  run_in_background=true,\n  load_skills=[\"git-master\"],\n  description=\"Mine all accessible contexts for missed requirements or background knowledge\",\n  prompt=\"\"\"\n\u003Creview_type>CONTEXT MINING - MISSED REQUIREMENTS & BACKGROUND\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS}\n\u003C\u002Fconstraints>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cbackground>\n{BACKGROUND}\n\u003C\u002Fbackground>\n\nYou are an investigator. Your mission: search every accessible information source to find context that should have informed this implementation but might have been missed. The question: \"Is there something we should have known but didn't?\"\n\nSOURCES TO SEARCH (use every available tool):\n\n1. **Git History** (ALWAYS search):\n   - `git log --oneline -20 -- {each changed file}` - recent changes and their reasons\n   - `git blame {critical sections}` - who wrote what and when\n   - `git log --all --grep=\"{keywords from goal}\"` - related commits\n   - Look for reverted commits, TODO\u002FFIXME\u002FHACK comments in history\n\n2. **GitHub** (if `gh` CLI available):\n   - `gh issue list --search \"{keywords}\"` - related open\u002Fclosed issues\n   - `gh pr list --search \"{keywords}\" --state all` - related PRs and their review comments\n   - Check if any issue is specifically linked to this work\n   - Look at review comments on past PRs touching these files\n\n3. **Communication Channels** (if MCP tools available):\n   - Slack: search for messages mentioning the feature, file names, or related keywords\n   - Notion: search for design docs, RFCs, ADRs related to this feature\n   - Discord: relevant discussions\n\n4. **Codebase Cross-References** (ALWAYS search):\n   - Files that import or reference the changed modules\n   - Tests that might need updating due to behavior changes\n   - Documentation (README, docs\u002F, comments) that references changed behavior\n   - Config files that might need corresponding updates\n   - Related features in the same domain\n\nWHAT TO LOOK FOR:\n\n- Requirements mentioned in issues\u002FPRs that the implementation misses\n- Past decisions explaining WHY code was written a certain way - and whether new changes respect those reasons\n- Related systems or features affected by these changes\n- Warnings from previous developers (PR review comments, inline TODOs, commit messages)\n- Migration or deprecation notes that affect the changed code\n- Design decisions documented outside the codebase (Notion, Slack, ADRs)\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Csources_searched>\n  - [SEARCHED\u002FSKIPPED] Source name - what was searched (or why it wasn't accessible)\n\u003C\u002Fsources_searched>\n\u003Cdiscovered_context>\n  For each discovery:\n  - Source: Where found (git commit abc123, GitHub issue #42, Slack message, etc.)\n  - Finding: What was found\n  - Relevance: How it relates to the current work\n  - Impact: [BLOCKING \u002F IMPORTANT \u002F FYI]\n\u003C\u002Fdiscovered_context>\n\u003Cmissed_requirements>Requirements the implementation should address but doesn't. Empty if none.\u003C\u002Fmissed_requirements>\n\u003Cblocking_issues>BLOCKING items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n## Phase 2: Wait & Collect\n\nAfter launching all 5 agents in one turn, wait for completions in bounded\ncycles. Do not treat a timeout, ack-only reply, or empty child result as\na PASS.\n\nAs each completes, collect via the Codex mapping above (`wait_agent`,\nthen the child's substantive final result). Store each verdict:\n\n| Agent | Verdict | Notes |\n|-------|---------|-------|\n| 1. Goal Verification | pending | - |\n| 2. QA Execution | pending | - |\n| 3. Code Quality | pending | - |\n| 4. Security | pending | - |\n| 5. Context Mining | pending | - |\n\nDo NOT deliver the final report until ALL 5 have completed.\nIf a lane remains silent after the reliability followup, record it as\ninconclusive and respawn a smaller reviewer\u002Fworker for that exact lane.\n\n---\n\n## Phase 3: Deliver Verdict\n\n\u003Cverdict_logic>\n\nALL 5 agents returned PASS → **REVIEW PASSED**\nANY agent returned FAIL → **REVIEW FAILED - criteria not met**\n\n\u003C\u002Fverdict_logic>\n\nCompile the final report in this format:\n\n```markdown\n# Review Work - Final Report\n\n## Overall Verdict: PASSED \u002F FAILED\n\n| # | Review Area | Agent Type | Verdict | Confidence |\n|---|------------|------------|---------|------------|\n| 1 | Goal & Constraint Verification | Oracle | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 2 | QA Execution | unspecified-high | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 3 | Code Quality | Oracle | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 4 | Security (supplementary) | Oracle | PASS\u002FFAIL | Severity |\n| 5 | Context Mining | unspecified-high | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n\n## Blocking Issues\n[Aggregated from all agents - deduplicated, prioritized]\n\n## Key Findings\n[Top 5-10 most important findings across all agents, grouped by theme]\n\n## Recommendations\n[If FAILED: exactly what to fix, in priority order]\n[If PASSED: non-blocking suggestions worth considering]\n```\n\nIf FAILED - be specific. The user should know exactly what to fix and in what order. No vague \"consider improving X\" - state the problem, the file, and the fix.\n\nIf PASSED - keep it short. Highlight any non-blocking suggestions, but don't turn a passing review into a lecture.\n","## ZCode Harness Tool Compatibility\n\nThis skill was ported from the Codex\u002FOpenCode OmO harness. ZCode does not provide `call_omo_agent`, `task`, `background_output`, `team_*`, `multi_agent_v1.spawn_agent`, or any Codex-only multi-agent tools. Translate every such example to ZCode's native tools:\n\n| Codex\u002FOpenCode example | ZCode tool to use |\n| --- | --- |\n| `call_omo_agent(subagent_type=\"explore\", ...)` | `Agent` tool with `subagent_type=\"Explore\"` and a self-contained `prompt` |\n| `call_omo_agent(subagent_type=\"librarian\", ...)` | `Agent` tool with `subagent_type=\"Explore\"` (research-focused prompt) |\n| `task(subagent_type=\"plan\", ...)` | `Agent` tool with `subagent_type=\"general-purpose\"` (planning-focused prompt) |\n| `task(subagent_type=\"oracle\", ...)` for final verification | `Agent` tool with `subagent_type=\"general-purpose\"` (rigorous-review prompt) |\n| `task(category=\"...\", ...)` for implementation or QA | `Agent` tool with `subagent_type=\"general-purpose\"` |\n| `multi_agent_v1.spawn_agent(...)` \u002F `spawn_agent(...)` | `Agent` tool with `subagent_type` and a self-contained `prompt` |\n| `background_output(task_id=\"...\")` | Only unstructured read-only research may use background mode; its completion arrives as a parent task notification, not a plugin hook |\n| `multi_agent_v1.wait_agent(...)` mailbox signals | No equivalent plugin tool; use returned foreground Agent results or the host's background research notifications |\n| `team_*(...)` | ZCode has no native team-of-threads API. Approximate with multiple parallel `Agent` calls (one per member) coordinated by the parent, or fall back to sequential subagents when isolation is unclear |\n\nZCode v3.14.3 loads enabled plugin agents from `agents\u002F`. Prefer namespaced roles such as `lazyz:explorer` and `lazyz:lazycodex-executor`; unambiguous bare names also work. Include all task context and ownership in the self-contained `prompt`. Roles inherit the user's active model; do not force an obsolete provider ID or copy Codex model\u002Feffort fields. If an older host lacks a role, describe it inline using the appropriate built-in agent.\n\nBefore every governed Agent or Task call, read the Dispatch schema in the installed [start-work skill](..\u002Fstart-work\u002FSKILL.md) and prepend exactly one LAZYCODEX_DISPATCH_V1 JSON line. General-purpose and known executor calls require this envelope; read-only Explore calls may remain unstructured. Use the same taskId across retries with a fresh runId, unchanged criteria, role, phase and budget. Planning\u002Fdocument work uses writer, code work implementer, review reviewer, and testing qa. Never rename a task to bypass its exhausted budget.\n\nGoverned calls must use `run_in_background: false` (or omit it). A background launch is not a completed result and has no matching plugin completion hook. Multiple independent foreground calls may be requested together; do not assume the host executes them concurrently. Read-only background research without a dispatch contract remains available.\n\nRequire each governed child to write the matching ReceiptV1 with current source identity, criterion invocations, zero exit codes and hashed regular artifacts, then return EVIDENCE_RECORDED; final-phase work requires FINAL_EVIDENCE_RECORDED. PostToolUse records failed evidence; Stop prevents unresolved completion while budget remains. An exhausted budget requires a blocker handoff, not an unsupported completion claim.\n\nUser authorization and verification policy take precedence over generic workflow instructions below. Do not auto-commit, push, merge, delete, deploy, install tools, call paid services, or run checks prohibited by the user. Record skipped checks honestly; do not fabricate a passing receipt. A task requiring a verified receipt remains unverified if the necessary checks are prohibited. ZCode Stop can continue at most three consecutive times, not indefinitely. Native Goal Mode, Memory and Wiki are host features; no callable API is assumed.\n\nFor work likely to take long, prefer multiple concurrent `Agent` calls in a single message over one giant blocking call. Treat each child's final message as the deliverable: if it lacks the required output, ack-only, or stops without producing evidence, fall back to a smaller re-scoped `Agent` call with the missing deliverable made explicit in the `prompt`.\n\n## Codex Subagent Reliability\n\nEvery `spawn_agent` message must be self-contained. Start with\n`TASK: \u003Cimperative assignment>`, then name `DELIVERABLE`, `SCOPE`, and\n`VERIFY`. State that it is an executable assignment, not a context\nhandoff. Role selection requires `agent_type`; `model` +\n`reasoning_effort` alone creates a default agent, not a reviewer or\nworker. Prefer `fork_turns: \"none\"` unless full history is truly\nrequired; paste only the review context that worker needs.\n\nPlan and reviewer agents may run for a long time; spawn them in the background, keep doing independent root work, and poll with short wait_agent cycles sized to the work. Never use a single long blocking wait for them, and never spin on tiny timeouts as a failure budget.\n\nTreat child status as a progress signal, not a timeout counter. For\nwork likely to exceed one wait cycle, require the child to send\n`WORKING: \u003Ctask> - \u003Ccurrent phase>` before long reading, testing, or\nreview passes, and `BLOCKED: \u003Creason>` only when it cannot progress.\nWhile any child is active, keep the parent visibly alive with active\nsubagent count, agent names, latest `WORKING:` phase, and whether the\nparent is waiting for mailbox updates. Track spawned agent names\nlocally. Use `wait_agent` for mailbox signals, not proof of completion.\nA timeout only means no new mailbox update arrived; after a timeout,\nrun a single `list_agents` check for the named child when you need\nreassurance. If it is running or its latest message is `WORKING:`,\ntreat it as alive. Do not use `list_agents` as a polling loop or status\nfeed; it can replay large payloads. Fallback only when the child is\ncompleted without the deliverable, ack-only after followup, explicitly\n`BLOCKED:`, or no longer running. Then mark that review lane\ninconclusive, do not count it as PASS or approval, close if safe, and\nrespawn a smaller `fork_turns: \"none\"` reviewer with the missing\ndeliverable.\n\n# Review Work - 5-Agent Parallel Review Orchestrator\n\nLaunch 5 specialized sub-agents in parallel to review completed implementation work from every angle. All 5 must pass for the review to pass. If even ONE fails, the review fails.\n\nWhen `review-work` is used as a final implementation, PR, or `$start-work`\ngate, it is blocking. A timeout, missing deliverable, ack-only response,\nexplicit `BLOCKED:`, or inconclusive lane is not a pass. Treat that lane as\nfailed, investigate the underlying uncertainty with the `debugging` skill when\nruntime behavior may be wrong, fix with evidence, and rerun the affected lane\nbefore claiming completion, creating or handing off a PR, or merging.\n\nA rejecting lane must name its blockers inline in its final message — each\nblocker cites the violated goal criterion or requirement plus an evidence\npointer. A bare REJECT\u002FFAIL token without findings is not a verdict; treat it\nas an inconclusive lane (one bounded respawn, then record it inconclusive with\nthat reason).\n\nWhen reviewing a PR or branch, collect diff, file contents, and verification\nresults from a dedicated review worktree attached to that branch. Never\ncheckout, test, or edit the review branch in the main worktree.\n\nReview evidence must be safe to share. Redact or mask secrets and sensitive\nuser data before including evidence in logs, PR bodies, or handoffs. Never\ninclude raw tokens, credentials, auth headers, cookies, API keys, env dumps,\nprivate logs, or PII; summarize with lengths, hashes, and short non-sensitive\nprefixes when identity is needed.\n\nThe 5 agents cover complementary concerns - together they form a comprehensive review that no single reviewer could match:\n\n| # | Agent | Type | Role | Focus Level |\n|---|-------|------|------|-------------|\n| 1 | Goal Verifier | Oracle | Did we build what was asked? | MAIN |\n| 2 | QA Executor | unspecified-high | Does it actually work? | MAIN |\n| 3 | Code Reviewer | Oracle | Is the code well-written? | MAIN |\n| 4 | Security Auditor | Oracle | Is it secure? | SUB |\n| 5 | Context Miner | unspecified-high | Did we miss any context? | MAIN |\n\n---\n\n## Phase 0: Gather Review Context\n\nBefore launching agents, collect these inputs. Extract from conversation history first - the user's original request, constraints discussed, and decisions made are usually already in the thread. Only ask if truly missing.\n\n\u003Crequired_inputs>\n\n- **GOAL**: The original objective. What was the user trying to achieve? Pull from the initial request in this conversation.\n- **CONSTRAINTS**: Rules, requirements, or limitations. Tech stack restrictions, performance targets, API contracts, design patterns to follow, backward compatibility needs.\n- **BACKGROUND**: Why this work was needed. Business context, user stories, related systems, prior decisions that informed the approach.\n- **CHANGED_FILES**: Auto-collect via `git diff --name-only HEAD~1` or against the appropriate base (branch point, specific commit).\n- **DIFF**: Auto-collect via `git diff HEAD~1` or against the appropriate base.\n- **FILE_CONTENTS**: Read the full content of each changed file (not just the diff). Oracle agents cannot read files - they need full context in the prompt.\n- **RUN_COMMAND**: How to start\u002Frun the application. Check `package.json` scripts, `Makefile`, `docker-compose.yml`, or ask the user.\n\n\u003C\u002Frequired_inputs>\n\n\n**NEVER CHECKOUT A PR BRANCH IN THE MAIN WORKTREE. ALWAYS CREATE A NEW GIT WORKTREE (`git worktree add`) AND WORK THERE. THIS PREVENTS CONTAMINATING THE USER'S WORKING DIRECTORY WITH UNRELATED BRANCH STATE.**\n\n**Auto-collection sequence:**\n\n```bash\n# 1. Get changed files\ngit diff --name-only HEAD~1  # or: git diff --name-only main...HEAD\n\n# 2. Get diff\ngit diff HEAD~1  # or: git diff main...HEAD\n\n# 3. Detect run command\n# Check package.json -> \"scripts.dev\" or \"scripts.start\"\n# Check Makefile -> default target\n# Check docker-compose.yml -> services\n```\n\nFor GOAL, CONSTRAINTS, BACKGROUND - review the full conversation history. The user's original message almost always contains the goal. Constraints often emerge during discussion. If anything critical is ambiguous, ask ONE focused question - not a checklist.\n\n---\n\n## Phase 1: Launch 5 Agents\n\nLaunch ALL 5 in a single turn. Every agent uses `run_in_background=true`. No sequential launches. No waiting between them.\n\n**Oracle agents receive everything in the prompt** (they cannot read files or run commands). Include DIFF + FILE_CONTENTS + all context directly in the prompt text.\n\n**unspecified-high agents are autonomous** - they can read files, run commands, and use tools. Give them goals and pointers, not raw content dumps.\n\n---\n\n### Agent 1: Goal & Constraint Verification (Oracle) - MAIN\n\nThis agent answers: \"Did we build exactly what was asked, within the rules we were given?\"\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Verify implementation against original goal and constraints\",\n  prompt=\"\"\"\n\u003Creview_type>GOAL & CONSTRAINT VERIFICATION\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL - paste the user's original request and any clarifications}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS - every rule, requirement, or limitation discussed}\n\u003C\u002Fconstraints>\n\n\u003Cbackground>\n{BACKGROUND - why this work was needed, broader context}\n\u003C\u002Fbackground>\n\n\u003Cchanged_files>\n{CHANGED_FILES - list of modified file paths}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of every changed file, clearly delimited per file}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF - the actual git diff}\n\u003C\u002Fdiff>\n\nReview whether this implementation correctly and completely achieves the stated goal within the given constraints. Be obsessively thorough - the point of this review is to catch what the implementer missed.\n\nREVIEW CHECKLIST:\n\n1. **Goal Completeness**: Break the goal into every sub-requirement (explicit AND implied). For each, mark ACHIEVED \u002F MISSED \u002F PARTIAL. Missing even one implied requirement that a reasonable engineer would have addressed = PARTIAL at minimum.\n\n2. **Constraint Compliance**: List every constraint. For each, verify compliance with specific code evidence. A constraint violated = automatic FAIL.\n\n3. **Requirement Gaps**: Requirements the user clearly wanted but didn't spell out. Things implied by the goal or background that a thoughtful engineer would have included.\n\n4. **Over-Engineering**: Anything added that wasn't requested - unnecessary abstractions, extra features, premature optimizations, speculative generality. Flag these as scope creep.\n\n5. **Edge Cases**: Given the goal, what inputs or scenarios would break this? Trace through at least 5 edge cases mentally.\n\n6. **Behavioral Correctness**: Walk through the code logic for 3+ representative scenarios. Does the code actually produce the expected behavior in each case?\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cgoal_breakdown>\n  For each sub-requirement:\n  - [ACHIEVED\u002FMISSED\u002FPARTIAL] Requirement description\n  - Evidence: specific code reference or gap\n\u003C\u002Fgoal_breakdown>\n\u003Cconstraint_compliance>\n  For each constraint:\n  - [ACHIEVED\u002FMISSED] Constraint description - evidence\n\u003C\u002Fconstraint_compliance>\n\u003Cfindings>\n  - [PASS\u002FFAIL\u002FWARN] Category: Description\n  - File: path (line range if applicable)\n  - Evidence: specific code or logic reference\n\u003C\u002Ffindings>\n\u003Cblocking_issues>Issues that MUST be fixed. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 2: QA via App Execution (unspecified-high) - MAIN\n\nThis agent answers: \"Does it actually work when you run it?\"\n\nThe QA agent follows a structured process: brainstorm scenarios exhaustively first, then self-review and augment, then create a task list, then execute systematically.\n\n```\ntask(\n  category=\"unspecified-high\",\n  run_in_background=true,\n  load_skills=[\"playwright\", \"dev-browser\"],\n  description=\"QA by actually running and using the application\",\n  prompt=\"\"\"\n\u003Creview_type>QA - HANDS-ON APP EXECUTION\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS}\n\u003C\u002Fconstraints>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Crun_command>\n{RUN_COMMAND - how to start the application, or \"unknown\" if not determined}\n\u003C\u002Frun_command>\n\nYou are a QA engineer. Your job is to RUN the application and verify it works through hands-on testing. You do not review code - you test behavior.\n\nMANDATORY PROCESS (follow in order):\n\n### Step 1: Scenario Brainstorm\n\nBefore touching the app, write down EVERY test scenario you can think of. Be exhaustive. Think about:\n\n- **Happy paths**: The primary use cases this implementation enables. What's the main thing the user wanted to do?\n- **Boundary conditions**: Empty inputs, maximum-length inputs, zero values, negative numbers, special characters, unicode, very large datasets.\n- **Error paths**: Invalid inputs, network failures, missing files, permission denied, timeout conditions.\n- **Regression scenarios**: Existing features that touch the same code paths. Things that worked before and must still work.\n- **State transitions**: What happens when you do things out of order? Rapid repeated actions? Concurrent usage?\n- **UX scenarios** (if applicable): Layout on different sizes, keyboard navigation, screen reader compatibility, loading states, error messages.\n- **Integration points**: Does this feature interact with external services, databases, or other modules? Test those boundaries.\n\nWrite each scenario as a one-liner with expected behavior. Aim for 15-30 scenarios minimum.\n\n### Step 2: Scenario Augmentation\n\nReview your scenario list with fresh eyes. For each scenario, ask:\n- \"What could go wrong here that I haven't considered?\"\n- \"What would a malicious or careless user do?\"\n- \"What environmental conditions could affect this?\" (disk full, slow network, expired tokens)\n\nAdd at least 5 more scenarios from this reflection. Group scenarios by priority: P0 (must pass), P1 (should pass), P2 (nice to pass).\n\n### Step 3: Create Task List\n\nConvert your augmented scenario list into a structured task list (use TaskCreate\u002FTaskUpdate or your todo system). Each task = one test scenario with:\n- Test name\n- Steps to execute\n- Expected result\n- Priority (P0\u002FP1\u002FP2)\n\n### Step 4: Execute Systematically\n\nWork through the task list in priority order (P0 first). For each test:\n\n1. Execute the test steps\n2. Record actual result\n3. Compare with expected result\n4. Mark PASS or FAIL\n5. If FAIL: capture evidence (screenshot, terminal output, error message)\n6. Mark the task complete\n\n**Execution guidance by app type:**\n- **Web app**: Use playwright\u002Fdev-browser to navigate, click, fill forms, verify visual output.\n- **CLI tool**: Run commands with various arguments, pipe inputs, check exit codes and output.\n- **Library\u002FSDK**: Write and execute a test script that imports and exercises the public API.\n- **Backend API**: Use curl\u002Fhttpie to hit endpoints with various payloads, verify response codes and bodies.\n- **Mobile\u002FDesktop**: If not directly runnable, write integration tests and execute them.\n\nIf the app cannot be started (build failure), that's an immediate FAIL - no need to continue.\n\n### Step 5: Compile Results\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cscenario_coverage>\n  Total scenarios: N\n  P0: X tested, Y passed\n  P1: X tested, Y passed\n  P2: X tested, Y passed\n\u003C\u002Fscenario_coverage>\n\u003Ctest_results>\n  For each test:\n  - [PASS\u002FFAIL] Test name (Priority)\n  - Steps: What you did\n  - Expected: What should happen\n  - Actual: What actually happened\n  - Evidence: Screenshot path or terminal output snippet (if FAIL)\n\u003C\u002Ftest_results>\n\u003Cblocking_issues>P0 or P1 failures only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 3: Code Quality Review (Oracle) - MAIN\n\nThis agent answers: \"Is the code well-written, maintainable, and consistent with the codebase?\"\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Review overall code quality, patterns, and architecture\",\n  prompt=\"\"\"\n\u003Creview_type>CODE QUALITY REVIEW\u003C\u002Freview_type>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of changed files AND neighboring files that show existing patterns}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF}\n\u003C\u002Fdiff>\n\n\u003Cbackground>\n{BACKGROUND}\n\u003C\u002Fbackground>\n\nYou are a senior staff engineer conducting a code review. Your standard: \"Would I approve this PR without comments?\"\n\nREVIEW DIMENSIONS (examine each):\n\n1. **Correctness**: Logic errors, off-by-one, null\u002Fundefined handling, race conditions, resource leaks, unhandled promise rejections.\n\n2. **Pattern Consistency**: Does new code follow the codebase's established patterns? Compare with the neighboring files provided. Introducing a new pattern where one already exists = finding.\n\n3. **Naming & Readability**: Clear variable\u002Ffunction\u002Ftype names? Self-documenting code? Would another engineer understand this without explanation?\n\n4. **Error Handling**: Errors properly caught, logged, and propagated? No empty catch blocks? No swallowed errors? User-facing errors are helpful?\n\n5. **Type Safety**: Any `as any`, `@ts-ignore`, `@ts-expect-error`? Proper generic usage? Correct type narrowing? (If TypeScript\u002Ftyped language)\n\n6. **Performance**: N+1 queries? Unnecessary re-renders? Blocking I\u002FO on hot paths? Memory leaks? Unbounded growth?\n\n7. **Abstraction Level**: Right level of abstraction? No copy-paste duplication? But also no premature over-abstraction?\n\n8. **Testing**: New behaviors covered by tests? Tests are meaningful, not just coverage padding? Test names describe scenarios?\n\n9. **API Design**: Public interfaces clean and consistent with existing APIs? Breaking changes flagged?\n\n10. **Tech Debt**: Does this introduce new tech debt? Or create coupling that will be painful to change?\n\nCategorize each finding by severity:\n- **CRITICAL**: Will cause bugs, data loss, or crashes in production\n- **MAJOR**: Significant quality issue that should be fixed before merge\n- **MINOR**: Improvement worth making but not blocking\n- **NITPICK**: Style preference, optional\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cfindings>\n  - [CRITICAL\u002FMAJOR\u002FMINOR\u002FNITPICK] Category: Description\n  - File: path (line range)\n  - Current: what the code does now\n  - Suggestion: how to improve\n\u003C\u002Ffindings>\n\u003Cblocking_issues>CRITICAL and MAJOR items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 4: Security Review (Oracle) - SUB\n\nThis agent answers: \"Are there security vulnerabilities in these changes?\"\n\nThis is supplementary - it focuses exclusively on security. It does NOT comment on code style, architecture, or functionality unless those directly create a security risk.\n\n```\ntask(\n  subagent_type=\"oracle\",\n  run_in_background=true,\n  load_skills=[],\n  description=\"Security-focused review of implementation changes\",\n  prompt=\"\"\"\n\u003Creview_type>SECURITY REVIEW (supplementary)\u003C\u002Freview_type>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cfile_contents>\n{FILE_CONTENTS - full content of changed files}\n\u003C\u002Ffile_contents>\n\n\u003Cdiff>\n{DIFF}\n\u003C\u002Fdiff>\n\nYou are a security engineer. Review this diff exclusively for security vulnerabilities and anti-patterns. Ignore code style, naming, architecture - unless it directly creates a security risk.\n\nSECURITY CHECKLIST:\n\n1. **Input Validation**: User inputs sanitized? SQL injection, XSS, command injection, SSRF vectors?\n2. **Auth & AuthZ**: Authentication checks where needed? Authorization verified for each action? Privilege escalation paths?\n3. **Secrets & Credentials**: Hardcoded secrets, API keys, tokens in code or config? Secrets in logs?\n4. **Data Exposure**: Sensitive data in logs? PII in error messages? Over-exposed API responses?\n5. **Dependencies**: New dependencies added? Known CVEs? Suspicious or unnecessary packages?\n6. **Cryptography**: Proper algorithms? No custom crypto? Secure random? Proper key management?\n7. **File & Path**: Path traversal? Unsafe file operations? Symlink following?\n8. **Network**: CORS configured correctly? Rate limiting? TLS enforced? Certificate validation?\n9. **Error Leakage**: Stack traces exposed to users? Internal details in error responses?\n10. **Supply Chain**: Lockfile updated consistently? Dependency pinning?\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cseverity>CRITICAL \u002F HIGH \u002F MEDIUM \u002F LOW \u002F NONE\u003C\u002Fseverity>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Cfindings>\n  - [CRITICAL\u002FHIGH\u002FMEDIUM\u002FLOW] Category: Description\n  - File: path (line range)\n  - Risk: What could an attacker do?\n  - Remediation: Specific fix\n\u003C\u002Ffindings>\n\u003Cblocking_issues>CRITICAL and HIGH items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n### Agent 5: Context Mining (unspecified-high) - MAIN\n\nThis agent answers: \"Did we miss any context that should have informed this implementation?\"\n\n```\ntask(\n  category=\"unspecified-high\",\n  run_in_background=true,\n  load_skills=[\"git-master\"],\n  description=\"Mine all accessible contexts for missed requirements or background knowledge\",\n  prompt=\"\"\"\n\u003Creview_type>CONTEXT MINING - MISSED REQUIREMENTS & BACKGROUND\u003C\u002Freview_type>\n\n\u003Coriginal_goal>\n{GOAL}\n\u003C\u002Foriginal_goal>\n\n\u003Cconstraints>\n{CONSTRAINTS}\n\u003C\u002Fconstraints>\n\n\u003Cchanged_files>\n{CHANGED_FILES}\n\u003C\u002Fchanged_files>\n\n\u003Cbackground>\n{BACKGROUND}\n\u003C\u002Fbackground>\n\nYou are an investigator. Your mission: search every accessible information source to find context that should have informed this implementation but might have been missed. The question: \"Is there something we should have known but didn't?\"\n\nSOURCES TO SEARCH (use every available tool):\n\n1. **Git History** (ALWAYS search):\n   - `git log --oneline -20 -- {each changed file}` - recent changes and their reasons\n   - `git blame {critical sections}` - who wrote what and when\n   - `git log --all --grep=\"{keywords from goal}\"` - related commits\n   - Look for reverted commits, TODO\u002FFIXME\u002FHACK comments in history\n\n2. **GitHub** (if `gh` CLI available):\n   - `gh issue list --search \"{keywords}\"` - related open\u002Fclosed issues\n   - `gh pr list --search \"{keywords}\" --state all` - related PRs and their review comments\n   - Check if any issue is specifically linked to this work\n   - Look at review comments on past PRs touching these files\n\n3. **Communication Channels** (if MCP tools available):\n   - Slack: search for messages mentioning the feature, file names, or related keywords\n   - Notion: search for design docs, RFCs, ADRs related to this feature\n   - Discord: relevant discussions\n\n4. **Codebase Cross-References** (ALWAYS search):\n   - Files that import or reference the changed modules\n   - Tests that might need updating due to behavior changes\n   - Documentation (README, docs\u002F, comments) that references changed behavior\n   - Config files that might need corresponding updates\n   - Related features in the same domain\n\nWHAT TO LOOK FOR:\n\n- Requirements mentioned in issues\u002FPRs that the implementation misses\n- Past decisions explaining WHY code was written a certain way - and whether new changes respect those reasons\n- Related systems or features affected by these changes\n- Warnings from previous developers (PR review comments, inline TODOs, commit messages)\n- Migration or deprecation notes that affect the changed code\n- Design decisions documented outside the codebase (Notion, Slack, ADRs)\n\nOUTPUT FORMAT:\n\u003Cverdict>PASS or FAIL\u003C\u002Fverdict>\n\u003Cconfidence>HIGH \u002F MEDIUM \u002F LOW\u003C\u002Fconfidence>\n\u003Csummary>1-3 sentence overall assessment\u003C\u002Fsummary>\n\u003Csources_searched>\n  - [SEARCHED\u002FSKIPPED] Source name - what was searched (or why it wasn't accessible)\n\u003C\u002Fsources_searched>\n\u003Cdiscovered_context>\n  For each discovery:\n  - Source: Where found (git commit abc123, GitHub issue #42, Slack message, etc.)\n  - Finding: What was found\n  - Relevance: How it relates to the current work\n  - Impact: [BLOCKING \u002F IMPORTANT \u002F FYI]\n\u003C\u002Fdiscovered_context>\n\u003Cmissed_requirements>Requirements the implementation should address but doesn't. Empty if none.\u003C\u002Fmissed_requirements>\n\u003Cblocking_issues>BLOCKING items only. Empty if PASS.\u003C\u002Fblocking_issues>\n\"\"\")\n```\n\n---\n\n## Phase 2: Wait & Collect\n\nAfter launching all 5 agents in one turn, wait for completions in bounded\ncycles. Do not treat a timeout, ack-only reply, or empty child result as\na PASS.\n\nAs each completes, collect via the Codex mapping above (`wait_agent`,\nthen the child's substantive final result). Store each verdict:\n\n| Agent | Verdict | Notes |\n|-------|---------|-------|\n| 1. Goal Verification | pending | - |\n| 2. QA Execution | pending | - |\n| 3. Code Quality | pending | - |\n| 4. Security | pending | - |\n| 5. Context Mining | pending | - |\n\nDo NOT deliver the final report until ALL 5 have completed.\nIf a lane remains silent after the reliability followup, record it as\ninconclusive and respawn a smaller reviewer\u002Fworker for that exact lane.\n\n---\n\n## Phase 3: Deliver Verdict\n\n\u003Cverdict_logic>\n\nALL 5 agents returned PASS → **REVIEW PASSED**\nANY agent returned FAIL → **REVIEW FAILED - criteria not met**\n\n\u003C\u002Fverdict_logic>\n\nCompile the final report in this format:\n\n```markdown\n# Review Work - Final Report\n\n## Overall Verdict: PASSED \u002F FAILED\n\n| # | Review Area | Agent Type | Verdict | Confidence |\n|---|------------|------------|---------|------------|\n| 1 | Goal & Constraint Verification | Oracle | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 2 | QA Execution | unspecified-high | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 3 | Code Quality | Oracle | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n| 4 | Security (supplementary) | Oracle | PASS\u002FFAIL | Severity |\n| 5 | Context Mining | unspecified-high | PASS\u002FFAIL | HIGH\u002FMED\u002FLOW |\n\n## Blocking Issues\n[Aggregated from all agents - deduplicated, prioritized]\n\n## Key Findings\n[Top 5-10 most important findings across all agents, grouped by theme]\n\n## Recommendations\n[If FAILED: exactly what to fix, in priority order]\n[If PASSED: non-blocking suggestions worth considering]\n```\n\nIf FAILED - be specific. The user should know exactly what to fix and in what order. No vague \"consider improving X\" - state the problem, the file, and the fix.\n\nIf PASSED - keep it short. Highlight any non-blocking suggestions, but don't turn a passing review into a lecture.\n",{"name":5,"description":7},"main","0a99b5b62ebd3726d465f7e319e2b8ec84c8aa6e","release\u002Flazyz-0.11.0-20261007\u002Fskills\u002Freview-work","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz","https:\u002F\u002Fgithub.com\u002Fyounghai","http:\u002Fbzengage.co","Bizmatrixx co-founder & CEO \r\n","2026-11-01 13:58:56",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add younghai\u002Flazyz --skill review-work",{"tool":38,"command":39},"bunx","bunx skills add younghai\u002Flazyz --skill review-work",{"tool":41,"command":42},"pnpm","pnpm dlx skills add younghai\u002Flazyz --skill review-work","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Frelease\u002Flazyz-0.11.0-20261007\u002Fskills\u002Freview-work\u002FSKILL.md",true,[46],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"agents\u002Fopenai.yaml","REFERENCE","参考文档",49,"https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Frelease\u002Flazyz-0.11.0-20261007\u002Fskills\u002Freview-work\u002Fagents\u002Fopenai.yaml",[53,60,67,74,81,87],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M78GVSC3PZBHQQ5HPRNQ3AMN","valid-skill","A valid skill for testing purposes.","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fvalid-skill\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M78GVRB79BN9BQ5MAZ43XJWX","unknown-field","A skill with an unknown frontmatter field.","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Funknown-field\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M78GVQ2319BDGJA2QXMM2DM3","nested-child","Nested skill should be ignored.","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Froot-skill-preferred\u002Fnested-child\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M78GVP5HC2C9B3VN8D8QWEFG","root-skill-preferred","Root skill should win.","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Froot-skill-preferred\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M78GVNAR8R5XF3W00182ZWZH","no-frontmatter","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fno-frontmatter\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M78GVK58G788T2TRHPCA8Y64","child-skill","A nested skill in a subdirectory.","https:\u002F\u002Fgithub.com\u002Fyounghai\u002Flazyz\u002Fblob\u002Fmain\u002Fxcode\u002Fagent\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fnested\u002Fchild-skill\u002FSKILL.md",[],[]]