enterprise-ai
Oracle Enterprise AI guidance for building, deploying, securing, estimating cost for, and integrating AI models, agents, RAG, Responses API workflow…
它会碰到什么
这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。
技能内容
OCI Enterprise AI Skills
Use this OCI subdomain as the front door for Oracle Enterprise AI work. Start here when the user goal is about the AI outcome rather than a specific Oracle service name.
How to Use This Domain
- Start with the user journey table below.
- Read only the smallest topic file that matches the task.
- Route database-owned implementation details to
db/features/instead of duplicating Select AI, vector search, orDBMS_VECTORguidance.
Directory Structure
oci/enterprise-ai/
├── models/
├── agent-workflows/
├── governance/
├── data/
├── cost/
├── integrations/
└── agents/
└── openai.yaml
User Journey Routing
| User Goal | Start With |
|-----------|------------|
| Choose or invoke OCI-hosted models, embeddings, rerank, endpoints, or dedicated serving | models/enterprise-ai-models.md |
| Import, fine-tune, host, or lifecycle-manage a custom model | models/custom-and-imported-models.md |
| Build an API-first agent with OpenAI-compatible OCI APIs | agent-workflows/responses-api-agents.md |
| Deploy a packaged agent runtime on OCI-managed infrastructure | agent-workflows/hosted-applications.md |
| Choose the right tool pattern: File Search, Code Interpreter, Function Calling, MCP, or SQL Search | agent-workflows/agent-tools.md |
| Ground answers in documents, vector stores, SQL, or Oracle Database vector search | data/rag-and-search.md |
| Estimate OCI Generative AI or OCI Generative AI Agents cost drivers | cost/cost-estimation.md |
| Secure model or agent access with IAM, private networking, OAuth, guardrails, or audit controls | governance/private-endpoints-and-governance.md |
| Connect Enterprise AI to Autonomous Database, APEX, Oracle Integration, Fusion, or OCI AI Services | integrations/oracle-platform-integrations.md |
Key Starting Points
oci/enterprise-ai/models/enterprise-ai-models.mdoci/enterprise-ai/models/custom-and-imported-models.mdoci/enterprise-ai/agent-workflows/responses-api-agents.mdoci/enterprise-ai/agent-workflows/hosted-applications.mdoci/enterprise-ai/agent-workflows/agent-tools.mdoci/enterprise-ai/data/rag-and-search.mdoci/enterprise-ai/cost/cost-estimation.mdoci/enterprise-ai/governance/private-endpoints-and-governance.md
Common Multi-Step Flows
| Task | Recommended Sequence |
|------|----------------------|
| Build a governed enterprise assistant | responses-api-agents -> agent-tools -> rag-and-search -> cost-estimation -> private-endpoints-and-governance |
| Move from model experimentation to production serving | enterprise-ai-models -> cost-estimation -> private-endpoints-and-governance |
| Bring a custom model into OCI serving | custom-and-imported-models -> enterprise-ai-models -> cost-estimation -> private-endpoints-and-governance |
| Deploy a packaged agent runtime | responses-api-agents -> hosted-applications -> cost-estimation -> private-endpoints-and-governance |
| Build RAG over files | rag-and-search -> agent-tools -> cost-estimation -> responses-api-agents |
| Build RAG over Oracle Database | rag-and-search -> db/features/vector-search.md -> db/features/dbms-vector.md -> db/features/select-ai.md when in-database generation is needed |
| Add AI to APEX or Autonomous Database apps | oracle-platform-integrations -> relevant db/features/ or apex/apexlang/ guidance |
| Design production access controls | private-endpoints-and-governance -> enterprise-ai-models or responses-api-agents depending on workload |
Scope Boundaries
- Keep cloud-side model, agent, tool, endpoint, hosting, and governance decisions in this domain.
- Use
db/features/select-ai.md,db/features/vector-search.md,db/features/dbms-vector.md, anddb/features/ai-profiles.mdfor Oracle Database-owned AI implementation. - Use
apex/apexlang/for APEX application artifact generation and APEX AI agent component templates. - Prefer official Oracle documentation for model availability, supported tools, IAM verbs, endpoint formats, and service limits because Enterprise AI services change frequently.
Sources
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/overview.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/home.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/models.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/fine-tune-models.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/imported-models.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/agents.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/governance.htm
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/oci-openai.htm
- https://www.oracle.com/artificial-intelligence/enterprise-ai/cost-estimator/
- https://docs.oracle.com/en-us/iaas/Content/generative-ai/calculate-cost.htm
想直接用这个技能?
本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。