ibm-self-serve-assets

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Use when writing, reviewing, or debugging Terraform/OpenTofu modules, tests, CI/CD pipelines, or state operations. Diagnoses failure modes (identity churn, secrets, blast radius, CI drift, state corruption) with version-aware guidance.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert skill for fetching, analyzing, optimizing, and securing IBM Maximo automation scripts with comprehensive best practices. Fetches scripts from Maximo environments via REST API and provides detailed optimization reports.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Scan a codebase for hardcoded secrets (API keys, passwords, tokens, cloud credentials, private keys), migrate them to HashiCorp Vault via the Vault MCP server, and replace each hardcoded value with a Vault SDK reference. Use when the user wants to audit, migrate, or remediate secrets in a repository.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for IBM watsonx.data OpenSearch vector search — covers k-NN index design with HNSW, IBM watsonx.ai embedding integration (ibm/slate-125m-english-rtrvr), document ingestion from IBM COS, hybrid search (BM25 + k-NN), score normalisation, and performance tuning. Generates production-ready Python 3.12…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Use when you need UFC insight pipeline best practices and coding standards — covers required code structure, utils usage, evidence objects, prestige ranking, heading generation, and the compliance checklist. Auto-activates when sdd-pipeline-builder needs the standards, or when the user asks about pipeline coding…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for building IBM RAG MCP servers (SSE and stdio transport) using the building-blocks mcp-server pattern — covers registerTool for ingest/retrieve/search operations, IBM watsonx.ai embedding calls, Milvus/OpenSearch connections, IBM COS source integration, SSE server deployment on IBM Code Engine, and…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for designing end-to-end IBM RAG pipelines — covers chunking strategy selection, IBM watsonx.ai embedding model choice (ibm/slate-125m-english-rtrvr), vector DB selection (IBM watsonx.data Milvus vs OpenSearch), hybrid search tuning, IBM COS document source, reranking with IBM watsonx.ai cross-encoder…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Add IBM watsonx.governance-backed runtime safety and quality guardrails to AI/RAG agents. Use when shipping LLM apps to production, designing 4-choke-point Pass/Flag/Block pipelines (input → retrieval → generation → output), picking metric sets from the 28-metric catalog, wiring guardrails into FastAPI / Flask /…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex needs its repo

Use when the user wants to create a new UFC insight pipeline from a GitHub issue or specification. Generates a spec.md, then builds the complete backend service, unit tests, registers it in the API, and opens a PR.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for enriching IBM watsonx.data Intelligence metadata to maximise Text2SQL accuracy — covers adding business descriptions, column synonyms, table relationships, and semantic hints via the DAI REST API, and using the metadataenrichment Python scripts in the building-blocks repository.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for evaluating and improving IBM watsonx.data Intelligence Text2SQL generated SQL quality — covers model selection (Granite vs Llama), dialect tuning, prompt engineering for SQL, SQL validation patterns, feedback loop design, error diagnosis, and metadata-driven accuracy improvement.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Use when the user wants to create a UDI flow, ingest documents into OpenSearch, set up a UDI data ingestion pipeline, run UDI document processing, configure UDI embeddings, or manage the UDI (Unstructured Data Integration) lifecycle on IBM Cloud watsonx — including registering COS and OpenSearch connections and…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Expert guidance for IBM watsonx.data zero-copy lakehouse configuration — covers Iceberg catalog setup, IBM COS and AWS S3 bucket registration via the watsonx.data REST API v2, Presto engine catalog association, Spark configuration, schema creation, and federated SQL queries. Generates production-ready Python 3.12…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex needs its repo

Build, review, explain, and extend the FactoryPulse Continuous RAG Streamhouse demo on Confluent Cloud. Use this skill for Kafka/Flink continuous knowledge ingestion, manufacturing exception-to-action flows, RAG freshness, Confluent configuration, API/UI changes, Bob-assisted development, and IBM Code Engine…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Measure, explain, and reduce the cost of watsonx Orchestrate (WXO) agents. Use when asked what an agent, a conversation, an evaluation run, or a use case costs; where the tokens go; how to turn platform traces or Langfuse data into dollars; how to set up the Langfuse integration and model pricing; how to compare cost…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Use when building, scaffolding, or replicating the IBM Cloudability + IBM Turbonomic Continuous FinOps React demo application with IBM Carbon g100 dark theme, complete API integrations, and multi-cloud optimization workflows.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

IBM Granite TTM time-series forecasting — predicts future values for cold-chain temperatures, demand, energy, stock prices, IoT sensors, or any numeric time series. Routes to the correct inference path (watsonx.ai API, local HuggingFace, or mock), generates service code, scaffolds a forecast dashboard, and supports…

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

Scaffold a self-contained TTM forecast dashboard.html — visualises 96-step predictions, risk score, time-to-breach, and inference path badge using Apache ECharts. Works by opening directly in a browser with no server required. Also covers wiring to a live forecast API and file placement conventions.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

TTM concurrent fleet forecasting — runs per-truck TTM predictions in parallel using ThreadPoolExecutor, handles timeouts and fallbacks, normalises cargo temperatures before building context windows, and logs governance payloads to IBM OpenScale in a fire-and-forget thread.

ibm-self-serve-assets/building-blocks

Skill Claude CodeCodex

TTM inference path selection and service code generation — covers watsonx.ai native API (Path 1), local HuggingFace pipeline (Path 2), and mock fallback (Path 3). Includes .env setup, IBM Cloud credentials, IAM token refresh, payload structure, output normalisation, Docker build, dependencies, and troubleshooting.

ibm-self-serve-assets/Bob-driven-Integration-for-Maximo-via-MCP

MCP server Claude CodeCodexCursor +2

MCP server "maximo-jobplan-server" as configured in ibm-self-serve-assets/Bob-driven-Integration-for-Maximo-via-MCP. Runs locally from the maximo-jobplan-server npm package.