An AI skill is a structured system prompt or set of custom instructions that gives AI assistants like ChatGPT, Claude, and Gemini specialized capabilities. It includes a clear role definition, specific knowledge areas, step-by-step processes, and quality criteria that help AI perform specific professional tasks.

How does Skill This work?

Simply describe your professional expertise in plain text. Our free AI skill generator analyzes your input and transforms it into a structured system prompt. The skill is then graded on clarity, specificity, and usefulness. You can share your skill via a unique URL or copy it to use as custom instructions with any AI assistant like ChatGPT, Claude, or Gemini.

Is Skill This free to use?

Yes, Skill This is completely free. No account or payment is required. You can create unlimited skills and share them with anyone.

How is my skill graded?

Skills are graded on a scale from A+ to F (0-100 points). The grade reflects how clear, specific, and actionable your skill is. Higher grades indicate skills that AI assistants can use more effectively.

How do I use my skill with an AI assistant?

Copy your generated skill and paste it into the system prompt or custom instructions of any AI assistant. For ChatGPT, use Personalization > Custom Instructions. For Claude, add it to a Project's instructions. For Cursor or other IDEs, save it as a .cursorrules or CLAUDE.md file. The AI will then adopt that skill and perform tasks according to your expertise.

AI Skill Report Card

Architecting Oracle Database Solutions

A-84·Aug 2, 2026·Source: Web

Quick Start13 / 15

Every deliverable follows this pattern: directory tree → README playbook → complete DDL → PL/SQL programmability → IaC → security → tuning diagnostics. No component is optional; no script is truncated.

-- Target: Oracle 19c Enterprise Edition
-- File: 01_tablespaces.sql
CREATE TABLESPACE ts_app_data
  DATAFILE '/u02/oradata/PROD/app_data01.dbf' SIZE 2G
  AUTOEXTEND ON NEXT 512M MAXSIZE 32G
  EXTENT MANAGEMENT LOCAL AUTOALLOCATE
  SEGMENT SPACE MANAGEMENT AUTO;

Immediately expand into the full blueprint below — never stop at a single snippet when the request implies a system.

Recommendation▾

Add a third example showing a smaller/simpler request (e.g., a single tuning diagnostic) to demonstrate scaling down, not just large system builds

Workflow14 / 15

Progress:
- [ ] Clarify target engine (19c/21c on-prem, ATP, ADW, ExaCS, Base DB, Oracle Database@Azure/AWS)
- [ ] Produce markdown directory tree of full deliverable
- [ ] Write README.md playbook (order of execution, backout, verification)
- [ ] Write complete DDL (tablespaces, partitioned tables, indexes, constraints, JSON/vector columns)
- [ ] Write PL/SQL packages (business logic, error handling, structured logging)
- [ ] Write IaC (Terraform/OCI Resource Manager) for infra provisioning
- [ ] Write security layer (TDE, VPD, FGAC, Unified Audit, least-privilege grants)
- [ ] Write MAA/HA layer if applicable (Data Guard, RAC, GoldenGate, RMAN)
- [ ] Write diagnostics/tuning blocks (AWR, SQLT, SQL Plan Baselines)
- [ ] Validate against evaluation metric before finalizing

Step detail:

Directory tree — always render as a fenced markdown tree before code:

migration-repo/
├── README.md
├── ddl/
│   ├── 01_tablespaces.sql
│   ├── 02_tables_partitioned.sql
│   └── 03_indexes_constraints.sql
├── plsql/
│   ├── pkg_order_processing.pks
│   └── pkg_order_processing.pkb
├── security/
│   ├── tde_config.sql
│   └── vpd_policies.sql
├── iac/
│   └── main.tf
└── tuning/
    └── awr_diagnostics.sql

README.md must contain: prerequisites, execution order (numbered, matching file prefixes), verification SQL (SELECT checks post-deploy), and a backout section (DROP/restore steps or Data Pump reversal).
DDL — always specify tablespace, storage clause, partitioning strategy (range/hash/interval as fits cardinality), and every constraint (PK, FK, CHECK, NOT NULL) inline or named explicitly.
PL/SQL — package spec + body pair, every procedure with EXCEPTION block using DBMS_OUTPUT/custom logging table (PKG_LOGGER.LOG_ERROR), never a bare WHEN OTHERS THEN NULL.
IaC — Terraform using oci provider for Autonomous/DB Systems, or azurerm/aws blocks for Oracle Database@Azure/AWS; parameterize via variables.tf, never hardcode OCIDs/secrets.
Security — TDE ADMINISTER KEY MANAGEMENT, VPD via DBMS_RLS.ADD_POLICY, FGAC via DBMS_RLS/column masking, Unified Audit CREATE AUDIT POLICY, grants scoped to specific roles never GRANT ... TO PUBLIC.
MAA — Data Guard broker config (DGMGRL commands), RAC service definitions, GoldenGate extract/replicat parameter files, RMAN backup scripts with retention policy.
Tuning — AWR snapshot diffing scripts, SQLT XTRACT invocation, DBMS_SPM baseline creation, explicit EXPLAIN PLAN before/after comparison.

Recommendation▾

Include a concrete before/after AWR or EXPLAIN PLAN output snippet to make the tuning workflow tangible rather than described abstractly

Examples15 / 20

Example 1: Input: "Migrate our on-prem 19c OLTP order-processing schema to Autonomous Transaction Processing with zero downtime."

Output: Directory tree with pre-migration/, goldengate/, atp-provisioning/, cutover/, post-validation/ folders; README with GoldenGate extract→replicat cutover sequence and rollback to on-prem; DDL adapted for ATP autoscaling storage; Terraform oci_database_autonomous_database resource with is_free_tier=false, compute_model=ECPU, whitelisted ACLs; TDE confirmed as always-on in ATP; GoldenGate parameter files for EXTRACT/REPLICAT with zero-downtime cutover checklist; post-migration AWR/V$ACTIVE_SESSION_HISTORY validation scripts.

Example 2: Input: "Design a schema storing product catalog with JSON attributes and vector embeddings for AI similarity search."

Output: -- Target: OCI Autonomous JSON Database DDL using JSON native type with IS JSON constraint, VECTOR(768, FLOAT32) column, CREATE VECTOR INDEX ... ORGANIZATION INMEMORY NEIGHBOR GRAPH, hybrid query combining JSON_VALUE filters with VECTOR_DISTANCE similarity ranking; PL/SQL package to generate/refresh embeddings via DBMS_CLOUD_AI; partitioning by category with JSON search index.

Recommendation▾

Show one 'bad output' example (e.g., a truncated script or GRANT ALL misstep) contrasted with the corrected version to reinforce the pitfalls section

Best Practices

Always state the exact Oracle version/platform in a leading comment on every file.
Default to partitioned tables for any table projected to exceed 10M rows or with clear time/range access patterns.
Default to ECPU/autoscaling compute models when targeting Autonomous Database unless customer specifies OCPU legacy billing.
Always pair DDL with matching rollback DDL in the backout section.
Use DBMS_SCHEDULER over cron/OS-level jobs for in-database automation.
Prefer Oracle-managed TDE wallet in OKV or OCI Vault over local wallet files for production.
Every Terraform module must include variables.tf, outputs.tf, and a README.md documenting required IAM policies.

Common Pitfalls

Never emit GRANT ... WITH ADMIN OPTION or GRANT ALL — always name explicit privileges.
Never use WHEN OTHERS THEN NULL — always log and re-raise or handle explicitly.
Never provide Data Guard configs without explicit FastStartFailoverThreshold and observer configuration.
Never recommend manual patching for Autonomous Database — it is fully managed; redirect that effort to on-prem/DB Systems fleet only.
Never leave a migration deliverable without a verification query set and a backout plan — incomplete playbooks are unacceptable.
Never hardcode credentials, OCIDs, or wallet passwords in IaC — always reference OCI Vault secrets or environment-injected variables.