用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/stevehuang0115/crewly --skill create-request命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | Create Request |
| description | Create a V3 Request from a user message that the orchestrator has determined to be actionable work. |
| version | 1.0.0 |
| category | task-management |
| skillType | claude-skill |
| assignableRoles | ["orchestrator","team-leader","developer","qa","tpm","designer","frontend-developer","backend-developer","fullstack-dev","qa-engineer","product-manager","architect","generalist"] |
| triggers | ["create request","log request","new request","track request"] |
| tags | ["v3","request","tracking"] |
| execution | {"type":"script","script":{"file":"execute.sh","interpreter":"bash","timeoutMs":15000}} |
Create a V3 Request entity to track actionable user work. This skill is called by the orchestrator (or any agent with semantic understanding) after determining that a user message represents real, actionable work.
Only call this skill when you determine a user message represents actionable work. Use your judgment as an LLM to classify intent.
DO create a Request for:
DO NOT create a Request for:
Use these to classify the complexity of the request:
| Level | Meaning | Example |
|---|---|---|
| L0 | Not actionable -- should NOT call this skill | "Thanks!", "Hello" |
| L1 | Simple, single-step task | "Fix the typo in README.md" |
| L2 | Multi-step, needs coordination or decomposition | "Build a new auth system with OAuth2 and tests" |
| L3 | Large project, should trigger Mission creation | "Rewrite the entire backend in Go with full test coverage" |
| Category | When to use |
|---|---|
| communication | Sending messages, notifications, status updates |
| query | Questions, lookups, information retrieval |
| code_change | Writing, modifying, or refactoring code |
| planning | Architecture decisions, roadmap work, design |
| debugging | Bug fixes, investigation, troubleshooting |
| deployment | Deploy, release, environment management |
| team_management | Creating teams, adding members, role changes |
| other | Anything that does not fit the above |
| Flag | Required | Default | Description |
|---|---|---|---|
--title | Yes | -- | Concise title summarizing the request |
--description | Yes | -- | Full description of what the user asked for |
--intent-level | Yes | -- | L0, L1, L2, or L3 |
--intent-category | Yes | -- | communication, query, code_change, planning, debugging, deployment, team_management, other |
--priority | No | normal | low, normal, high, urgent |
--source-message-id | No | agent-self-{session}-{epoch} | Original message ID (for dedup). When omitted, the skill synthesizes agent-self-{CREWLY_SESSION_NAME}-{epoch} so agent-context callers (e.g. TL self-filing a child Request) don't 400 on the backend's required-field check. Issue #458. |
--needs-mission | No | false | Whether this L3 request should trigger Mission creation |
# Simple bug fix
bash execute.sh --title "Fix auth test failure" \
--description "The test in auth.service.test.ts is failing due to a missing mock" \
--intent-level L1 --intent-category debugging \
--source-message-id "msg_abc123"
# Complex multi-step request
bash execute.sh --title "Build OAuth2 authentication system" \
--description "Create a complete OAuth2 auth system with Google and GitHub providers, including tests and documentation" \
--intent-level L2 --intent-category code_change \
--priority high --source-message-id "msg_def456"
# Large project needing Mission
bash execute.sh --title "Rewrite backend in Go" \
--description "Full backend rewrite from Node.js to Go with all existing functionality preserved" \
--intent-level L3 --intent-category code_change \
--priority high --needs-mission true --source-message-id "msg_ghi789"
bash execute.sh '{"title":"Fix auth test","description":"Fix failing test","intentLevel":"L1","intentCategory":"debugging"}'
JSON response with the created Request object including:
id -- Unique Request IDtitle -- The title providedstatus -- "open" (initial status)intentLevel -- The classified intent levelintentCategory -- The classified intent category