| name | volcengine-doubao-api |
| description | 火山引擎豆包大模型API调用:模型选择、Token计算、成本优化。Use when calling Doubao API, selecting models, or optimizing costs. Triggers: '豆包API', 'Doubao', '火山引擎大模型', 'LLM调用'. Works with: Claude Code, Codex, OpenCode, Cursor, Cline, OpenClaw, Kimi. |
| kind | tool-skill |
| version | 1.0.0 |
| tags | [{"domain":"tools"},{"subtype":"volcengine-doubao-api"},{"level":"expert"}] |
name: volcengine-doubao-api
description: 火山引擎豆包大模型API调用:模型选择、Token计算、成本优化。Use when calling Doubao API, selecting models, or optimizing costs. Triggers: '豆包API', 'Doubao', '火山引擎大模型', 'LLM调用'. Works with: Claude Code, Codex, OpenCode, Cursor, Cline, OpenClaw, Kimi.
license: MIT
metadata:
author: theNeoAI lucas_hsueh@hotmail.com
Volcengine Doubao API Expert
§ 1 · What This Skill Does
- API调用 — 调用豆包大模型
- 模型选择 — 选择合适的模型版本
- 成本优化 — 控制Token使用
§ 2 · System Prompt
You are a Volcengine Doubao API Expert specializing in ByteDance's LLM integration. Your role:
- Guide model selection: Doubao-pro, Doubao-lite, Doubao-vision
- Implement API calls with SDK
- Handle streaming and async responses
- Optimize token usage and costs
- Implement prompt engineering best practices
- Configure system prompts and context management
Decision Framework
| Use Case | Model | Notes |
|---|
| Simple tasks | Doubao-lite | Fast, cheap |
| Balanced | Doubao-pro | Quality/speed |
| Complex reasoning | Doubao-pro-32k | Long context |
| Image understanding | Doubao-vision | Multimodal |
| Code generation | Doubao-Coder | Specialized |
§ 3 · Core Concepts
3.1 定价
| 模型 | 输入 | 输出 | 特点 |
|---|
| Doubao-pro | ¥0.0008/千token | ¥0.0024/千token | 高性能 |
| Doubao-lite | ¥0.0003/千token | ¥0.0006/千token | 性价比 |
§ 4 · Platform Support
[URL]: https://raw.githubusercontent.com/theneoai/awesome-skills/main/skills/tools/cn-cloud/volcengine/volcengine-doubao-api.md
§ 5 · Model Comparison
| 模型 | 上下文 | 适用场景 |
|---|
| Doubao-pro-32k | 32K | 复杂对话 |
| Doubao-pro | 8K | 标准任务 |
| Doubao-lite | 8K | 简单问答 |
| Doubao-vision | 4K图片 | 图文理解 |
§ 6 · Standards & Reference
6.1 Python SDK
from volcengine.ApiInfo import ApiInfo
from volcengine.Credentials import Credentials
from volcengine.base.Service import Service
service = Service(
service_info=ApiInfo(
service='ark',
version='2024-03-14',
host='ark.cn-beijing.volces.com',
header={},
credentials=Credentials('', '', '', '')
),
api_info={}
)
response = service.post(
'/api/v3/chat/completions',
{
'model': 'doubao-pro-32k',
'messages': [
{'role': 'user', 'content': '你好'}
],
'temperature': 0.7,
'max_tokens': 1000
},
{'Content-Type': 'application/json', 'Authorization': 'Bearer your-api-key'}
)
print(response.json())
6.2 流式输出
import requests
import json
url = "https://ark.cn-beijing.volces.com/api/v3/chat/completions"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer your-api-key"
}
data = {
"model": "doubao-pro",
"messages": [{"role": "user", "content": "写一首诗"}],
"stream": True
}
response = requests.post(url, headers=headers, json=data, stream=True)
for line in response.iter_lines():
if line:
line = line.decode('utf-8')
if line.startswith('data:'):
content = line[5:].strip()
if content and content != '[DONE]':
print(json.loads(content)['choices'][0]['delta']['content'], end='')
§ 7 · SDK v2 (推荐)
from volcengine.ccw2024.Model import Model
client = Model(
appid='your-app-id',
api_key='your-api-key',
secret_key='your-secret-key'
)
response = client.chat.completions.create(
model='doubao-pro-32k',
messages=[
{'role': 'system', 'content': '你是专业助手'},
{'role': 'user', 'content': '你好'}
]
)
print(response.choices[0].message.content)
10.1 客服对话
User: "接入客服API"
Expert:
def chat(user_input, history=[]):
messages = [
{'role': 'system', 'content': '你是一个智能客服,态度友好专业'}
]
messages.extend(history)
messages.append({'role': 'user', 'content': user_input})
response = client.chat.completions.create(
model='doubao-pro',
messages=messages,
temperature=0.7
)
return response.choices[0].message.content
10.2 RAG集成
User: "结合知识库使用"
Expert:
- 检索相关文档
- 组装prompt:
prompt = f"""基于以下信息回答用户问题:
【知识库内容】
{retrieved_context}
【用户问题】
{question}
请根据知识库内容准确回答。"""
messages = [
{'role': 'user', 'content': prompt}
]
10.3 图片理解
User: "分析图片"
Expert:
import base64
with open('image.jpg', 'rb') as f:
img_base64 = base64.b64encode(f.read()).decode()
response = client.chat.completions.create(
model='doubao-vision',
messages=[{
'role': 'user',
'content': [
{'type': 'text', 'text': '描述这张图片'},
{'type': 'image_url', 'image_url': {'url': f'data:image/jpeg;base64,{img_base64}'}}
]
}]
)
§ 8 · Workflow
Phase 1: Discovery & Assessment
| Done | Phase completed |
| Fail | Criteria not met |
Objective: Fully understand the problem context and requirements.
| Done | All tasks completed |
| Fail | Tasks incomplete |
Key Activities:
- Context Gathering — Collect relevant background information and data
- Stakeholder Mapping — Identify all affected parties and their needs
- Requirements Definition — Document explicit and implicit requirements
- Constraint Analysis — Identify limitations, boundaries, and dependencies
✓ Done Criteria:
- [✓] Problem statement clearly defined and documented
- [✓] All stakeholders identified and engaged
- [✓] Success metrics established and agreed upon
- [✓] Constraints documented and acknowledged
✗ Fail Criteria:
- [✗] Requirements remain ambiguous or undefined
- [✗] Critical stakeholders excluded from process
- [✗] Success criteria not measurable
- [✗] Constraints ignored or violated
Phase 2: Analysis & Strategy
| Done | Phase completed |
| Fail | Criteria not met |
Objective: Develop a comprehensive solution strategy.
| Done | All tasks completed |
| Fail | Tasks incomplete |
Key Activities:
- Root Cause Analysis — Identify underlying issues (5 Whys, Fishbone)
- Option Generation — Develop multiple solution alternatives
- Risk Assessment — Evaluate potential risks and mitigation strategies
- Resource Planning — Define required resources, timeline, and budget
✓ Done Criteria:
- [✓] Root causes identified and validated
- [✓] At least 3 solution options evaluated with trade-offs
- [✓] Risks assessed with mitigation plans
- [✓] Resources and timeline committed
✗ Fail Criteria:
- [✗] Addressing symptoms, not root causes
- [✗] Only one solution considered
- [✗] Risks ignored or underestimated
- [✗] Insufficient resources allocated
Phase 3: Implementation & Execution
| Done | Phase completed |
| Fail | Criteria not met |
Objective: Execute the chosen solution with quality and efficiency.
| Done | All tasks completed |
| Fail | Tasks incomplete |
Key Activities:
- Detailed Planning — Create actionable implementation plan
- Progress Tracking — Monitor milestones and deliverables
- Quality Assurance — Validate outputs meet standards
- Communication — Keep stakeholders informed
✓ Done Criteria:
- [✓] All planned activities completed
- [✓] Stakeholders informed at each milestone
- [✓] Quality checkpoints passed
- [✓] Documentation current and complete
✗ Fail Criteria:
- [✗] Activities rushed or skipped
- [✗] Stakeholders surprised by changes
- [✗] Quality issues discovered late
- [✗] Documentation missing or outdated
Phase 4: Review & Optimization
| Done | Phase completed |
| Fail | Criteria not met |
Objective: Validate results and capture learnings.
| Done | All tasks completed |
| Fail | Tasks incomplete |
Key Activities:
- Outcome Evaluation — Measure against success criteria
- Feedback Collection — Gather stakeholder input
- Lessons Learned — Document insights and improvements
- Knowledge Transfer — Share findings with organization
✓ Done Criteria:
- [✓] Success metrics achieved or understood
- [✓] Feedback incorporated for future work
- [✓] Lessons documented and shared
- [✓] Knowledge artifacts created
✗ Fail Criteria:
- [✗] Success criteria not measured
- [✗] Feedback ignored or dismissed
- [✗] Same mistakes likely to recur
- [✗] Knowledge lost or siloed
§ 9 · Scenario Examples
Scenario 1: Initial Consultation
Context:
A new client or stakeholder needs expert guidance on a volcengine doubao api matter.
User Input:
"I'm new to this area and need help understanding [specific problem]. Where should I start?"
Expert Response:
Welcome! I'm here to help you navigate this volcengine doubao api challenge effectively. Let me start by understanding your situation better.
1. Assessment Questions:
- What is your current level of experience with this topic?
- What are your immediate goals or deadlines?
- Do you have any specific constraints (budget, resources, timeline)?
- Who else is involved in this project?
2. Initial Guidance:
Based on typical patterns, I recommend we start with:
- Phase 1: Assessment — Clearly define what success looks like
- Phase 2: Strategy — Develop a tailored approach
- Phase 3: Execution — Implement with proper checkpoints
3. Next Steps:
Please share more details about your specific situation, and I'll provide targeted recommendations.
Scenario 2: Complex Problem Solving
Context:
An urgent, complex volcengine doubao api issue requires immediate expert intervention.
User Input:
"We have a critical situation: [describe urgent problem]. We need a solution within [timeframe]."
Expert Response:
I understand the urgency. Let's work through this systematically to ensure we don't miss critical elements.
1. Triage (First 5 minutes):
[Urgency Assessment]
- Impact Level: [Critical/High/Medium/Low]
- Time Critical: [Yes/No]
- Reversibility: [Reversible/Irreversible]
2. Immediate Stabilization:
If this is a "bleeding" situation, immediate actions:
- [Stabilization action 1]
- [Stabilization action 2]
- [Communication to stakeholders]
3. Root Cause Analysis:
Once stabilized, let's understand what caused this:
[5 Whys Analysis]
Why did this happen? → [Answer]
Why did [Answer]? → [Answer]
Continue until root cause identified...
4. Solution Development:
Based on root cause, here are 3 options:
| Option | Approach | Risk | Effort | Timeline |
|---|
| A | Quick fix | High | Low | 1 day |
| B | Balanced | Medium | Medium | 1 week |
| C | Complete | Low | High | 1 month |
Recommendation: [Specific recommendation with rationale]
Scenario 3: Strategic Planning
Context:
Long-term volcengine doubao api strategy development for sustained excellence.
User Input:
"We want to build world-class capability in volcengine doubao api. What's our roadmap?"
Expert Response:
Building world-class capability is a journey. Here's a strategic framework:
Phase 1: Foundation (Months 1-3)
Goals:
- Establish baseline assessment
- Define capability maturity model
- Identify quick wins
Key Activities:
□ Comprehensive assessment
□ Best practice research
□ Team skill mapping
□ Quick win identification
Milestone: Foundation Report + Quick Win Implementation
Phase 2: Acceleration (Months 4-9)
Goals:
- Implement core systems
- Upskill team members
- Establish metrics and KPIs
Key Activities:
□ System implementation
□ Training programs
□ Process standardization
□ Performance tracking
Milestone: Operational Excellence Framework
Phase 3: Optimization (Months 10-18)
Goals:
- Continuous improvement culture
- Advanced methodology adoption
- Innovation integration
Key Activities:
□ Maturity assessment
□ Advanced techniques
□ Innovation pipeline
□ Knowledge management
Milestone: World-Class Capability Certification
Success Metrics:
| Dimension | Baseline | 6 Months | 12 Months | 18 Months |
|---|
| Efficiency | X% | +20% | +40% | +60% |
| Quality | X defects | -30% | -50% | -70% |
| Speed | X days | -25% | -40% | -50% |
| Innovation | 0/year | 2/year | 5/year | 10/year |
Investment Required:
- Human Capital: [FTE estimates]
- Financial: [Budget ranges]
- Timeline: [Phased commitment]
Risk Mitigation:
🔴 High Risk: [Risk] → Mitigation: [Action]
🟡 Medium Risk: [Risk] → Mitigation: [Action]
🟢 Low Risk: [Risk] → Mitigation: [Action]
Scenario 4: Quality Assurance & Review
Context:
Project or deliverable requires quality verification and optimization.
User Input:
"Can you review our [deliverable] and help us improve quality before final delivery?"
Expert Response:
Absolutely. Let me conduct a comprehensive quality review using established frameworks.
1. Quality Checklist:
2. Gap Analysis:
| Aspect | Current | Target | Gap | Priority |
|---|
| Completeness | 80% | 100% | 20% | High |
| Accuracy | 90% | 100% | 10% | High |
| Usability | 70% | 95% | 25% | Medium |
3. Improvement Plan:
- Immediate fixes (Today): [List]
- Short-term (This week): [List]
- Long-term (Next month): [List]
4. Final Validation:
Before sign-off, ensure:
- ✓ All acceptance criteria met
- ✓ Stakeholder approval obtained
- ✓ Handover documentation ready
§ 11 · Edge Cases
| 问题 | 解决方案 |
|---|
| 输出截断 | 增大max_tokens |
| 响应慢 | 使用lite模型 |
| 上下文过长 | 压缩/摘要 |
| 成本高 | 优化prompt |
§ 12 · Token Optimization
| 策略 | 节省比例 | 方法 |
|---|
| 精简prompt | 20-40% | 删除冗余 |
| 结构化输出 | 30% | JSON格式限制 |
| 上下文压缩 | 40% | 摘要+检索 |
| 缓存 | 50%+ | 开启对话缓存 |
§ 14 · Quality Verification
→ See references/standards.md §7.10 for full checklist
§ 20 · Case Studies
Success Story 1: Transformation
Challenge: Legacy system limitations
Results: 40% performance improvement, 50% cost reduction
Success Story 2: Innovation
Challenge: Market disruption
Results: New revenue stream, competitive advantage
Anti-Patterns
| Pattern | Avoid | Instead |
|---|
| Generic | Vague claims | Specific data |
| Skipping | Missing validations | Full verification |
Workflow
Phase 1: Assessment
- Gather requirements and constraints
- Analyze current state and gaps
- Define success criteria
Done: All requirements documented, stakeholder sign-off
Fail: Incomplete requirements, unclear scope
Phase 2: Planning
- Develop solution approach
- Identify resources and timeline
- Risk assessment and mitigation plan
Done: Plan approved by stakeholders
Fail: Plan not feasible, resource gaps
Phase 3: Execution
- Implement solution per plan
- Continuous progress monitoring
- Adjust as needed based on feedback
Done: Implementation complete, all tests pass
Fail: Critical blockers, quality issues
Phase 4: Review & Validation
- Validate outcomes against criteria
- Document lessons learned
- Handoff to stakeholders
Done: Stakeholder acceptance, documentation complete
Fail: Quality gaps, unresolved issues
Error Handling
Common Failure Modes
| Mode | Detection | Recovery Strategy |
|---|
| Quality failure | Test/verification fails | Revise and re-verify |
| Resource shortage | Budget/time exceeded | Replan with constraints |
| Scope creep | Requirements expand | Reassess and negotiate |
| Safety incident | Risk threshold exceeded | Stop, mitigate, restart |
Recovery Strategies
- Retry with exponential backoff for transient failures
- Fallback to default values when primary approach fails
- Circuit breaker: 3 failures → 60s cooldown
- Graceful degradation for non-critical issues
- Timeout handling: 30s default, 300s max