القائمة
Research Oracle/AI topics with confidentiality-aware synthesis, OCI patterns, and structured knowledge capture for consulting engagements
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Build and validate OCI draw.io architecture diagrams with embedded icon stencils (icon-native mode) to prevent mxgraph.oci fallback rendering issues like red placeholder blocks. Use for any OCI diagram creation or refactor in draw.io.
Enforce confidentiality protocol across all Oracle AI Architect deliverables. Pre-delivery audit, codename enforcement, and content sanitization. This skill has VETO power — if it fails, delivery is blocked.
Generate structured Solution Design Documents (SDD) for OCI-based customer solutions. Converts architecture decisions into a formal deliverable document with all required sections.
Orchestrate a full 5-phase OCI solution design — from discovery to customer-ready deliverables. Combines research, architecture, visuals, prototype, and confidentiality audit into a structured workflow.
Portable, scalable methodology for building AI agent systems using the AGENT Blueprint framework — validated at enterprise and personal scale
Expert guidance for OCI Dedicated AI Clusters — private LLM hosting, fine-tuning, sizing, and cost optimisation for enterprise deployments
| name | oracle-research |
| description | Research Oracle/AI topics with confidentiality-aware synthesis, OCI patterns, and structured knowledge capture for consulting engagements |
| version | 2.0.0 |
| platform | ["claude-code","cline","cursor","roocode"] |
| activation | {"claude_code":"/oracle-research","cline":"@skills/oracle-research/SKILL.md","cursor":"@skills/oracle-research/SKILL.md"} |
This skill enables confidential research for Oracle consulting work, with built-in OCI knowledge and abstraction protocols.
research/topics/"OCI [service] best practices 2026"
"Oracle Cloud [service] architecture patterns"
"[service] vs [competitor] enterprise comparison"
"enterprise GenAI architecture 2026"
"LLM [use case] production deployment"
"RAG implementation [industry] enterprise"
"OKE [feature] production patterns"
"Kubernetes [pattern] enterprise scale"
"service mesh [option] OCI integration"
"[industry] AI use cases 2026"
"[industry] cloud transformation patterns"
"Oracle [industry] solutions"
| Service | Use Case | Key Feature |
|---|---|---|
| Compute VMs | General workloads | Flexible shapes |
| Bare Metal | High performance | Dedicated hardware |
| GPU Instances | AI/ML training | A100, A10, L40S |
| Container Instances | Serverless containers | No cluster overhead |
| Service | Use Case | Key Feature |
|---|---|---|
| OCI Generative AI | LLM inference | Cohere, Meta models |
| AI Services | Pre-built AI | Vision, Language, Speech |
| Data Science | ML development | Notebooks, jobs, models |
| Dedicated AI Clusters | Large-scale AI | Reserved GPU capacity |
| Service | Use Case | Key Feature |
|---|---|---|
| Autonomous DB | Self-managing | Auto-tuning, patching |
| MySQL HeatWave | Analytics | In-memory acceleration |
| NoSQL | Key-value/JSON | Massive scale, low latency |
| Object Storage | Data lake | S3-compatible, tiers |
| Service | Use Case | Key Feature |
|---|---|---|
| OKE | Kubernetes | Managed control plane |
| Virtual Nodes | Serverless K8s | No node management |
| Container Registry | Image storage | Scanning, signing |
| Real | Abstracted |
|---|---|
| Customer name | Codename (K, V, E) or "the customer" |
| Specific amount | Range or "significant" |
| Exact date | Quarter (Q1, Q2) |
| Specific region | "primary region" |
# [Topic]
**Date:** YYYY-MM-DD
**Type:** Quick / Standard / Deep Dive
**Project:** K / V / E / General
## Summary
[2-3 sentences]
## Key Findings
1. [Finding with source]
2. [Finding with source]
## Application
[How this applies to current work]
## Sources
- [URL 1]
- [URL 2]
# [Topic]: OCI vs Alternatives
**Date:** YYYY-MM-DD
**Context:** [Why comparing]
## Comparison Matrix
| Criteria | OCI | Alternative |
|----------|-----|-------------|
| [Criterion] | [OCI position] | [Alt position] |
## OCI Advantages
- [Advantage 1]
## Areas to Address
- [Gap 1]
## Positioning Guidance
[How to discuss with customers]
ALWAYS check these before recommending custom solutions:
| Repository | URL | When to Check |
|---|---|---|
| OCI AI Blueprints | https://github.com/oracle-quickstart/oci-ai-blueprints | GenAI on OKE, vLLM, Llama Stack, fine-tuning |
| AI Accelerator Packs | docs/ai_accelerator_packs | Route optimization, video search, enterprise reasoning |
| technology-engineering | https://github.com/oracle-devrel/technology-engineering | Solution patterns, architecture guides |
| oracle-quickstart | https://github.com/oracle-quickstart | Terraform modules, reference architectures |
Customer needs GenAI deployment?
├── Standard LLM inference → OCI AI Blueprints (vLLM)
├── Complete agentic stack → OCI AI Blueprints (Llama Stack)
├── Fine-tuning → OCI AI Blueprints (LoRA Fine-Tuning)
├── Route optimization → AI Accelerator Pack (cuOPT)
├── Video analysis → AI Accelerator Pack (VSS)
├── Enterprise chat agent → AI Accelerator Pack (AI-Q)
└── Unique requirements → Custom with oracle-devrel patterns
When research relates to K, V, or E:
research/projects/[CODENAME]/Weekly, review research/topics/ and create:
research/synthesis/week-YYYY-WW.md
Research findings can become:
This skill is automatically loaded when research context is detected.
To use this skill in Cline, reference it at the start of your message:
@skills/oracle-research/SKILL.md
Research OCI Generative AI pricing options for a high-volume document processing use case.
Or in a .clinerules workflow:
## Research Tasks
When asked to research Oracle or OCI topics, load @skills/oracle-research/SKILL.md and follow the Standard Research protocol. Save output to research/topics/.
Triggers: Oracle research, OCI comparison, architecture patterns, GenAI use cases