oracle

Build, configure, scaffold, and deploy OCI Functions from a local machine using a dependency-first, Fn-context-guided flow with argv-safe mutation execution, nonce-scoped confirmations, and a canonical lowercase machine-readable contract. Use when asked to validate/install Fn, OCI CLI, and Docker, configure or create Fn contexts, validate OCI session and Docker registry auth from Fn context values, choose or create a Functions application, scaffold Java, Python, or Node functions, and deploy them. Network creation is supported only as an explicit advanced branch.

Diagnose OCI Functions setup, control-plane, deployment, invocation, and observability issues using a deterministic triage workflow. Use when asked to classify Fn CLI errors, deploy or invoke failures, OCIR auth and registry issues, app-limit symptoms, logs, tracing, metrics, or limits. Diagnose first, prefer read-only checks, and do not execute fixes unless the user explicitly asks.

Oracle Cloud Infrastructure guidance for designing, operating, and troubleshooting OCI services, including OCI Kubernetes Engine (OKE), OCI Internet of Things Platform, OCI Functions deployment and troubleshooting, and Enterprise AI workflows for OCI Generative AI models, Responses API agents, RAG, cost estimation, governance, private endpoints, hosted agentic applications, and Oracle platform integrations. Use when the user asks about OKE cluster design, Terraform or Resource Manager planning, OKE incident troubleshooting, Generic VNIC Attachment, Multus, pod networking, node pools, add-ons, ingress, load balancers, OCIR image pulls, Workload Identity, Kubernetes workloads on OCI, OCI IoT domains or digital twins, device publish flows, OCI Functions setup, deployment, invocation, or troubleshooting, OCI Generative AI, Enterprise AI Models, Enterprise AI Agents, governed GenAI applications, agentic workflows, RAG on Oracle Cloud, or OCI Generative AI pricing.

Public APEXlang router with deterministic local-context discovery and compact machine-readable contracts.

Oracle APEX skills for Oracle APEX application development.

Explore, create, and troubleshoot Oracle Cloud Infrastructure Internet of Things Platform resources, including domains, digital twin models, adapters, instances, relationships, and device publish flows. Use when requests involve OCI IoT bootstrap, digital twin inspection, safe lifecycle operations, or translating OCI IoT docs and samples into concrete commands.

Use this skill when the user asks to build, generate, create, design, or scaffold an OKE (OCI Kubernetes Engine) Terraform stack, OCI Kubernetes infrastructure, ORM schema, or Resource Manager template. Trigger phrases include "build an OKE stack", "create OKE Terraform", "generate ORM schema", "deploy OKE cluster", "OKE infrastructure", "terraform-oci-oke", or any request to design OCI Kubernetes infrastructure with Terraform. Do not use it for active incident RCA, GVA secondary VNIC node-pool creation, or Multus pod manifest validation; use the troubleshooting, GVA, or multihome skills for those surfaces.

Use this skill when the user asks to enable, deploy, or configure Generic VNIC Attachment (GVA) for OCI Kubernetes Engine (OKE), create node pools with secondary VNIC profiles, review update plans for existing GVA node pools, map Application Resources to workloads, or explain GVA functionality, constraints, and scheduling behavior. Do not use it for general OKE incident RCA or for deploying Multus test pods after a GVA node pool already exists; use oke-troubleshooter or oke-multihome-deployer for those surfaces.

Review pull requests with the `library-new-request` label in graalvm-reachability-metadata. Use when asked to review or triage a PR that adds metadata and tests for a new library, including PRs titled like `[GenAI] Add support for com.fasterxml:classmate:1.5.1 using gpt-5.4`. Focus on catching scaffold-only tests, version-pinned or package-bypassing tests, and PRs that push more than one library.

Review pull requests with the `library-update-request` label in graalvm-reachability-metadata. Use when asked to review or triage a PR that updates metadata/tests for an existing library version, especially generated PRs titled like `[GenAI] Improve coverage for group:artifact:version using gpt-5.5`. Focus on preserving dynamic-access coverage percentage between versions, verifying reporter-requested metadata is present and exercised by tests, and applying the relevant test/scope rules from new-library reviews without requiring a minimum new-library dynamic-access coverage threshold.

Fix missing GraalVM reachability metadata for a specific library. Use when the user asks to fix or add metadata because required metadata is missing and tests or native-image runs fail with missing metadata errors.

Review pull requests with the `fixes-java-run-fail` label in graalvm-reachability-metadata. Use when asked to review or triage a PR described as `fixes-java-run-fail` or one that fixes Java/JVM runtime test failures for an existing library version update. Focus on validating the Java runtime fix, keeping the diff scoped, and ensuring the dynamic-access coverage percentage does not drop by more than 20 percentage points between the previously tested version and the new version.

Review pull requests with the `fixes-javac-fail` label in graalvm-reachability-metadata. Use when asked to review or triage a PR that fixes Java compilation failures for an existing library version update. Focus on validating the compile fix, keeping the diff scoped, and ensuring the dynamic-access coverage percentage does not drop by more than 20 percentage points between the previously tested version and the new version.

Review pull requests with the `fixes-native-image-run-fail` label in graalvm-reachability-metadata. Use when asked to review or triage a PR that fixes a native-image runtime failure for an existing library version update. Focus on validating the runtime/native fix, keeping the diff scoped, and ensuring the dynamic-access coverage percentage does not drop by more than 20 percentage points between the previously tested version and the new version.

Fix graalvm-reachability-metadata PRs where final validateIndexFiles against current master fails because metadata index.json tested versions are in the wrong metadata-version bucket, duplicated across buckets, or stale after another index-changing PR landed. Use during PR review when an index-changing PR must be repaired and pushed instead of labeled human-intervention.

Review automated pull requests with the `library-bulk-update` label in graalvm-reachability-metadata. Use when asked to review or triage a PR that bumps tested versions for existing libraries, including approve vs close decisions, CI checks, diff-scope validation, and verification of `source-code-url`, `test-code-url`, `documentation-url`, and `repository-url` changes.

NetSuite Intelligence skill — teaches AI the correct tool selection order, output formatting, domain knowledge, multi-subsidiary and currency handling, and SuiteQL safety checklist for any AI + NetSuite AI Service Connector session.

Use this skill when the user needs NetSuite-based finance analysis, Director of Financial Analysis support, period-close guidance, variance review, reconciliation review, cash reporting, or executive-ready financial narrative generation. Apply it to requests involving income statements, balance sheets, cash flow, budget versus actuals, AR or AP aging, month-end close, SOX-oriented finance checks, journal entry review, and board or CFO reporting. Prefer it when the request clearly involves NetSuite financial data interpretation or finance operations, but do not use it for general NetSuite development or non-finance administration tasks.

Platform-agnostic OWASP secure coding practices with JavaScript/Node.js patterns and NetSuite SuiteScript examples. Covers Open Worldwide Application Security Project (OWASP) Top 10 (2021), output encoding, injection prevention, CSP headers, file security, API hardening, AI agent security, DRY security patterns, and 48+ security pitfalls with GOOD/BAD code templates.

Generate enterprise-grade documentation for NetSuite SDF projects. Analyze scripts, object XML files, `manifest.xml`, and SuiteQL queries to produce README.md, architecture diagrams (Mermaid/ASCII), deployment guides, and troubleshooting tables. Can integrate with post-deployment documentation workflows when automation (for example, hooks) is available.

Use when generating or reviewing NetSuite SDF permission configurations such as customrole XML, script deployment permissions, permkey values, permlevel choices, run-as role design, and least-privilege access. Confirms exact ADMI_ / LIST_ / REGT_ / REPO_ / TRAN_ permission IDs, distinguishes standard permissions from customrecord_* script IDs, and validates permissions against bundled NetSuite reference data.

Comprehensive NetSuite SDF best practices based on the SAFE Guide (12 principles + appendices). Generates Object XML for all 14 script types, enforces governance limits, security patterns, and defensive coding. Includes N/cache, N/query, concurrency limits, OAuth 2.0 guidance, legacy TBA guardrails, CustomTool runtime patterns, REST Web Services (2026.1 features), and 140+ documented pitfalls. Essential for SuiteApp and Account Customization development.

Interactive learning system for NetSuite SDF development. Features six modes (learn, review, explain, annotate, quiz, and final) with SAFE Guide integration. Produces compliance-reviewed learning documentation. Learn topics like governance, N/cache, and security directly from SAFE Guide principles. Generate quizzes from code or SAFE Guide content.

SuiteScript records and fields reference. Look up field IDs, types, required status, and search capabilities for all 272 NetSuite record types. Use this when building SuiteScript to ensure correct field usage.

Windows-specific command guidance. Use when working on Windows and a required CLI tool, especially gdev-cli, is not available directly in PowerShell or cmd.

Create or update GraalPy third-party package compatibility patches under graalpython/lib-graalpython/patches, including PyPI source preparation, rebasing existing patches, metadata.toml updates, license checks, version-range validation, and verify_patches.py validation.

Analyze recent GraalPy benchmark regressions on `master` as part of the weekly rota. Use when asked to analyze benchmarks for rota.

Analyze current GraalPy periodic job failures for ROTA. Use when asked to triage, summarize, or plan work for current periodic job failures, starting from scripts/rota_ci_failures.py output, validating linked Jira issues, inspecting logs, forming hypotheses, reproduction commands, and implementation order.

Mirror an external GitHub pull request into the internal GraalPython Bitbucket review flow, including OCA label checks, Jira creation or reuse, preserving PR commits, pre-commit cleanup, and handoff to the shared GraalPython Bitbucket PR flow for PR creation, Graal Bot tasks, gates, and fixes.

Create or continue a GraalPython Bitbucket pull request and drive it through Graal Bot tasks, gate start, gate monitoring, failure investigation, fixes, pushes, and gate reruns. Use after a branch is ready for internal GraalPython review, or when an automation command has already created the PR and the remaining work is task cleanup and gate follow-up.

Run the GraalPy ROTA import update workflow. Use when asked to refresh imports, create the standard Graal import update pull request, inspect generated commits, and hand off to the shared GraalPython Bitbucket PR flow for tasks, gates, and failure fixes.

Check gate status for a Bitbucket PR by resolving the PR head commit, finding the gate merge commit, inspecting builds on that merge commit, and summarizing root-cause failures with actionable next steps.

Complete CLI reference for the ADS AQUA command-line interface (ads aqua). Covers all model, deployment, evaluation, and fine-tuning commands with full parameter documentation. Triggered when user asks about CLI commands, wants to run AQUA operations from terminal, or needs command syntax.

Deploy LLM models on OCI using AI Quick Actions (AQUA) - single model, multi-model, stacked (LoRA), with GPU shape selection, vLLM configuration, streaming, and tool calling. Triggered when user wants to deploy, update, or manage model deployments.

Evaluate LLM model quality using BERTScore, ROUGE, Perplexity, and Text Readability metrics on OCI AI Quick Actions (AQUA). Covers dataset preparation, evaluation job creation, and report interpretation. Triggered when user wants to evaluate or benchmark a model.

Fine-tune LLM models using LoRA on OCI AI Quick Actions (AQUA). Covers dataset preparation (instruction, conversational, multimodal, tokenized formats), hyperparameter tuning, distributed training, and training metrics. Triggered when user wants to fine-tune or customize a model.

Set up Prometheus and Grafana monitoring for AQUA vLLM model deployments on OCI. Covers the signing proxy, container registry setup, OCI Container Instance deployment, and PromQL dashboards. Triggered when user wants to monitor LLM deployments, view TTFT/latency/throughput metrics, or set up observability for AQUA.

Register, list, get, and manage LLM models in OCI AI Quick Actions (AQUA) using the ADS SDK. Triggered when user wants to import models from HuggingFace or Object Storage, browse available models, or manage model catalog entries.

Diagnose and fix OCI AI Quick Actions (AQUA) issues including deployment failures, OOM errors, authorization problems, capacity issues, container errors, and policy misconfigurations. Triggered when user encounters errors or needs help debugging AQUA workflows.

OCI Data Science service patterns including Jobs, Pipelines, Model Catalog, authentication, and the ADS SDK beyond AQUA. Triggered when user asks about OCI Data Science Jobs, ML Pipelines, model artifacts, conda environments, or general ADS SDK usage.

Build and troubleshoot GraalVM Native Image applications. Use this skill to build Java applications with GraalVM Native Image, configure CLI options, or resolve build or runtime issues.

Build GraalVM native images using the native-maven-plugin (org.graalvm.buildtools). Use this skill to build Java applications with Maven, configure pom.xml native image settings, run native tests, collect metadata, or resolve build or runtime issues.

Build GraalVM native images using Gradle Native Build Tools. Use this skill to build Java applications with Gradle, configure native-image build.gradle settings, or resolve build or runtime issues.

Create Open Agent Spec / PyAgentSpec agents or flows from business requirements, natural-language product ideas, or integration intents. Use when Codex must turn plain-English intent into an Agent Spec artifact, refine an existing Agent Spec artifact, choose between Agent and Flow designs, define LLM/tool/input/output contracts, add MCP tools or toolboxes, or validate an Agent Spec artifact without relying on a local agent-spec checkout.