Use when working with Oracle Database, Oracle SQL, PL/SQL, sqlplus, cx_Oracle, oracledb, ORDS, OCI drivers, Oracle containers, schema migrations, security, vectors, or performance tuning.

Use when editing Mojo code, .mojo files, fire emoji files, SIMD kernels, Python-Mojo interop, GIL-free parallelism, C FFI, hatch-mojo build hooks, or packaging Mojo extensions.

Use when editing Makefile or GNUmakefile, adding development targets, wiring uv commands, defining .PHONY rules, creating self-documenting help, or fixing Make recipe safety.

Use when implementing inter-process communication, shared memory regions, SPSC or MPMC ring buffers, zero-copy data transfer, platform synchronization primitives, or process notification mechanisms.

Use when installing, updating, packaging, or troubleshooting Flow integrations across Claude Code, Gemini CLI, Codex CLI, OpenCode, Cursor, VS Code/Copilot, OpenClaw, or Google Antigravity.

Use when building Inertia.js apps, editing createInertiaApp, server-side routed SPAs, Inertia protocol responses, page components, shared props, or Litestar/Inertia integrations.

Use when editing hx-* attributes, building HTMX hypermedia flows, returning partial HTML responses, setting HTMX response headers, or rendering server-side .html templates.

Use when deploying ASGI, WSGI, or RSGI apps with Granian, editing granian CLI commands, worker or thread settings, SSL, HTTP/2, backpressure, or replacing uvicorn for production.

Use when working with GKE, kubectl, Kubernetes manifests, k8s directories, Helm charts, node pools, workload identity, cluster scaling, GPU nodes, database sidecars, or GKE troubleshooting.

Use when managing Google Cloud resources, editing .gcloudignore or app.yaml, scripting gcloud commands, configuring IAM, service accounts, Cloud Storage, Pub/Sub, BigQuery, Vertex AI, or GCP services.

Use when a repository has .agents, when the user asks for Flow lifecycle routing, Beads-backed task memory, spec-first planning, TDD implementation, sync/status, review, finish, archive, or /flow:* help.

Use when syncing Beads state to markdown, checking Flow status, refreshing context docs, validating task markers, or reporting ready/blocked Flow work.

Use when initializing Flow in a repo, configuring .agents, installing or checking Beads bd, setting local-only sync policy, or creating first project context files.

Use when drafting PRDs, researching, planning, refining, revising, or creating .agents/specs/<flow_id>/spec.md worksheets for Flow.

Use when implementing Flow tasks from Beads or spec.md, claiming ready work, applying TDD, recording task notes, committing, and syncing after task state changes.

Use when reviewing, finishing, archiving, reverting, validating, or cleaning up Flow work after implementation or phase completion.

Use when writing DuckDB SQL, using .duckdb files, duckdb imports, analytical queries, CSV/Parquet/JSON ingestion, ETL pipelines, extensions, client APIs, configuration, or performance tuning.

Use when editing Dockerfile, Containerfile-like Docker syntax, docker-compose.yml, docker-compose.yaml, .dockerignore, multi-stage builds, BuildKit cache mounts, Compose services, or image optimization.

Use when generating documentation, writing API docs, documenting modules or components, creating README content, building reference guides, or documenting many files with explicit coverage tracking.

Use when editing Dishka dependency injection code, Provider, Scope, Container, FromDishka, Inject, DI scopes, providers, testing containers, or Litestar/FastAPI Dishka integrations.

Use when reviewing PRs, evaluating design proposals, assessing technical plans, stress-testing assumptions, looking for overlooked failure modes, or adding pushback before a decision.

Use when a problem resists quick answers, debugging stalls, analysis feels shallow, confidence is low, hypotheses are competing, reasoning loops repeat, or a hard problem needs evidence tracking.

Use when editing C++ files, .cpp, .hpp, .cc, .hh, .cxx, CMakeLists.txt, modern C++ APIs, resource ownership, error handling, concurrency, build systems, or native extension code.

Use when comparing architectural choices, technology options, build-vs-buy decisions, feature proposals, high-impact tradeoffs, multi-team decisions, hard-to-reverse choices, or risk-heavy alternatives.