Agent skills for OpenAI Agents SDK

Design patterns: stack organization, component catalogs, inheritance, configuration composition, version management, layered configuration

CI/CD: GitHub Actions, Spacelift, Atlantis, `atmos describe affected` for change detection

Go templates: Sprig/Gomplate functions, Atmos's own collectKeys, atmos.Component, atmos.GomplateDatasource, atmos.Store, template…

Helmfile orchestration: sync/apply/destroy/diff, Kubernetes deployments, varfile generation, EKS integration, source management

Introspection & Querying: describe/list commands, config filtering, workspace introspection, dependency graphs, YQ integration…

JSON Schema for Atmos: stack-manifest and atmos.yaml config schemas, IDE auto-completion, validate stacks/schema/config, SchemaStore…

Packer orchestration: init/build/validate/inspect/output, machine image building, template management, source management

Atmos root configuration: atmos.yaml discovery, precedence, deep merging, base_path, imports, minimal bootstrap, and routing to narrower…

Terraform and OpenTofu orchestration: plan/apply/deploy, workspace management, backend config, varfile generation, authentication, binary…

Toolchain management: declarative dependencies, automatic installs, install/exec/search/env commands, Aqua registry integration, version…

Workflow automation: native step types, multi-step workflows, parallel/matrix/wait/container/emulator steps, when: conditions (CEL)…

Crash-safe, all-or-nothing export of measurement data — temp-file→flush→fsync→rename atomic writes for CSV/JSON/text, path-yield variants…

Use when creating atomic-level UI components like buttons, inputs, labels, and icons.

Use when applying Atomic Design methodology to organize UI components into quarks, atoms, molecules, organisms, templates, and pages.

Use when creating page layouts without real content.

Attach, explore, and persist DuckDB (.duckdb) or SQLite (.sqlite) database files across multi-turn sessions.

USAP agent skill for Attack Surface Management.

Build comprehensive attack trees to visualize threat paths.

Hunt LLM/AI feature bugs — prompt injection, indirect injection, exfiltration via tool-use/markdown, ASCII smuggling, agentic AI security…

Attend the loop-lane human queue: escalated items and untriaged intake in one view, driven to resolution.

Diagnose and fix the top Attio REST API errors by HTTP status code.