flow-code

Internal pipeline engine. Manages the entire pipeline (brainstorm, plan, plan-review, work, impl-review, close) via flowctl phase commands. Invoked by /flow-code:go.

Use when exploring requirements before planning. Pressure-tests ideas, generates approaches, and outputs a requirements doc for /flow-code:plan. Supports --auto mode for AI self-interview (no human input needed).

Use when exploring requirements before planning. Pressure-tests ideas, generates approaches, and outputs a requirements doc for /flow-code:plan.

Use when planning features or designing implementation. Triggers on /flow-code:plan with text descriptions or Flow IDs.

Use when reviewing code changes — self-review in Worker Phase 6, impl-review, or PR review. Applies five-axis scoring with severity labels.

Use when implementing a plan or working through a spec. Triggers on /flow-code:work with Flow IDs.

Use when implementing authentication, authorization, OAuth, JWT, RBAC, or session management. Covers token lifecycle, permission models, and secure session handling.

Use when adding caching layers — HTTP cache, CDN, Redis, in-memory, or application-level. Covers cache invalidation, TTL strategy, and cache-aside patterns.

Use when writing Dockerfiles, docker-compose configs, or Kubernetes manifests. Covers image optimization, multi-stage builds, health checks, and security hardening.

Use when designing schemas, writing migrations, optimizing queries, or working with ORMs. Covers migration safety, indexing strategy, and query optimization.

Use when writing ADRs, API docs, READMEs, changelogs, or any technical documentation. Covers the 'document the why' principle, ADR templates, and doc-as-code workflow.

Use when designing error handling strategy, implementing retry logic, circuit breakers, or graceful degradation. Covers error classification, recovery patterns, and user-facing error UX.

Skill discovery guide. Use at session start or when unsure which skill to apply. Shows the flowchart for selecting the right flow-code skill.

Use when adding multi-language support, locale formatting, RTL layouts, or any internationalization/localization work.

Use when designing service boundaries, inter-service communication, data ownership, or distributed system patterns. Covers service mesh, saga, event-driven, and decomposition strategies.

Use when setting up dashboards, SLOs/SLIs, alerting rules, or on-call runbooks. Covers the monitoring hierarchy from health checks to incident response.

Use when adding logging, tracing, metrics, or health endpoints. Covers structured logging, distributed tracing, metric collection, and alerting integration.

Use when implementing WebSocket, SSE, long-polling, or any real-time communication. Covers protocol selection, connection management, reconnection, and scaling.

Use when designing state architecture for frontend or full-stack apps. Covers state classification, tool selection, server state patterns, and common pitfalls.

Use when building or modifying user-facing interfaces, creating components, implementing layouts, managing state, or when UI output needs to look production-quality rather than AI-generated

Use when writing code that handles user input, authentication, authorization, secrets, file uploads, or external data. Enforces OWASP Top 10 prevention and three-tier security boundaries.

Use when implementing with test-first methodology in Worker Phase 4, fixing bugs with Prove-It Pattern, or establishing test coverage strategy.

Use when implementing features in Worker Phase 5. Enforces vertical slicing, incremental commits, scope discipline, and the Implement-Test-Verify-Commit cycle.

Use when reducing code complexity, removing dead code, refactoring for clarity, or when auto-improve targets simplification. Preserves behavior exactly while improving readability.

Use when doing comprehensive plan review from multiple perspectives. Chains CEO/product, engineering, design, and DX reviews sequentially. Triggers on /flow-code:autoplan, 'full review', 'multi-perspective review'.

Full autopilot pipeline: brainstorm → plan → plan-review → work → impl-review → close. Zero human input from idea to PR.

Use when user wants to autonomously optimize code quality, performance, security, or test coverage. Triggers on /flow-code:auto-improve, 'auto improve', or 'run experiments on'.

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes

Use when asking 'what's blocking what', 'execution order', 'dependency graph', 'what order should epics run', 'critical path', or 'which epics can run in parallel'.

Use when working on Django projects — architecture, DRF APIs, ORM patterns, security, testing, and deployment verification. Triggers on Django-related tasks.

Use when all epic tasks are done and need a final review before closing. Triggers on /flow-code:epic-review.

Use when you want to review code or plans with an external model (ChatGPT, Claude web, etc.). Triggers on /flow-code:export-context.

Use when reviewing code changes, PRs, or implementations. Triggers on /flow-code:impl-review.

Use when user wants to flesh out a spec, refine requirements, or clarify a feature before building. Triggers on /flow-code:interview with Flow IDs or file paths.

Use when user asks 'how is the run going', 'loop status', 'ralph status', 'what iteration', or /flow-code:loop-status.

Use when user says 'map this codebase', 'document the architecture', 'create codebase map', or /flow-code:map.

Use when reviewing Flow epic specs or design docs. Triggers on /flow-code:plan-review.

Use when assessing codebase readiness for agents or production. Triggers on /flow-code:prime, 'assess this codebase', or 'is this repo ready'.

Internal guidance for composing Codex/GPT review and task prompts. Loaded automatically by worker agents and review skills when building prompts for cross-model review.

Use when user wants to set up the Ralph autonomous loop. Triggers on /flow-code:ralph-init.