메뉴
Review code changes for correctness, regressions, security risks, and missing tests. Use whenever the user asks for a review or before merging non-trivial work.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Create and update robust Daycare tasks (task_create/task_update + task_trigger_add) and permanent agents (create_permanent_agent) with minimal Python orchestration, explicit parameter schemas, reproducible exec/process usage, and performance-first skip() behavior. Use when users ask for cron/webhook automation, recurring checks, scheduled reports, permanent agents, or resilient background task workflows.
Create reusable Daycare voice agents with the hidden-by-default `voice_agent_create` tool. Use when users want a new voice persona, a realtime call prompt, client-side voice tools, or a saved voice agent that appears in the app Voice screen.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Daycare's capabilities with specialized knowledge, workflows, or tool integrations.
Understand what Daycare tasks are, how cron and webhook triggers work, and when to use each. Use when users ask about tasks, scheduling, triggers, or want to inspect existing tasks.
Friendship mechanics for daycare agents. Use when the user wants to add friends, manage friend requests, send messages to friends, share or unshare subusers with friends, or understand what nametags are.
Work with user-scoped PostgreSQL databases (PGlite). Use when the user asks to store structured data, create tables, query rows, track records, build a database, or manage any tabular/relational information. Covers database creation, additive schema design, CRUD operations, and SQL queries.
| name | code-review |
| description | Review code changes for correctness, regressions, security risks, and missing tests. Use whenever the user asks for a review or before merging non-trivial work. |
Treat review as bug-finding, not prose generation.
Check these first:
Always lead with findings. Keep them ordered by severity.
1. High: [short title] — [why it is wrong]
File: /absolute/path/to/file.ts:123
2. Medium: [short title] — [risk or missing case]
File: /absolute/path/to/file.ts:45
Open questions:
- [only if intent is unclear]
Rules: