felixti

Implement OpenSpec change using Ralph Loop - parallel specialized agents with fresh context per iteration. Use when user wants faster implementation with domain experts working in parallel, or when the change spans multiple domains (backend, frontend, database).

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.

Archive multiple completed changes at once. Use when the user wants to clean up several completed changes.

Continue working on an OpenSpec change by creating the next artifact. Use when the user wants to progress their change, create the next artifact, or continue their workflow.

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

Fast-forward through OpenSpec artifact creation. Use when the user wants to quickly create all artifacts needed for implementation without stepping through each one individually.

Start a new OpenSpec change using the artifact workflow. Creates a structured change with proposal, specs, design, and tasks. Use when the user wants to create a new feature, fix, or modification with a structured step-by-step approach.

OpenTelemetry GenAI semantic conventions for instrumenting LLM, agent, and RAG applications. Use this skill whenever someone is instrumenting generative AI code with OpenTelemetry, choosing gen_ai.* span/metric/event attributes, building observability for LLM calls, agent pipelines, embeddings, retrieval, or tool execution. Also use when reviewing or debugging GenAI telemetry, setting up AI observability with Dynatrace/Jaeger/Grafana, or working with OpenLLMetry. Covers spans, metrics, events, and provider-specific conventions for OpenAI, Anthropic, AWS Bedrock, and Azure AI Inference.

OpenTelemetry observability patterns: traces, metrics, logs, context propagation, OTLP export, Collector pipelines, and troubleshooting

Run verification pipeline after making code changes. Use this skill when the user asks to verify, check, validate, or test their changes, or when you've finished implementing something and need to ensure it works. Also triggers on phrases like "run the checks", "make sure it passes", "verify this works".

Build (write/code) an opencode plugin that adds an OAuth login for a custom or self-hosted model provider — primarily the OAuth 2.0 device-code grant (RFC 8628) for headless setups (SSH, VPS, Docker, CI) — with Microsoft Entra ID (Azure AD) as the worked example. Use this when the user wants to create their own login in `opencode auth login` / the `/connect` picker by implementing the plugin `auth` hook (methods + authorize + a loader): running a device-code or browser OAuth flow, injecting a bearer token, refreshing the access token (e.g. via client.auth.set), or wiring opencode up to Azure OpenAI, Entra ID, Okta, Keycloak, Auth0, Google, Ping, or any OIDC/OAuth provider or internal model gateway. Trigger this even when phrased loosely, e.g. "add Entra ID login to opencode", "opencode custom provider OAuth", "authenticate opencode with our SSO", "opencode device-code auth plugin", or "let opencode log in to our internal model gateway". Do NOT use this for: signing into opencode's own built-in providers (Open