continue-framework

Review the latest work unit with fresh, unnarrowed eyes and surface any problem it contains as follow-ups. Use after a work-unit commit, or when the user runs /agent-review.

Continue working on a story or task by reading progress.md, executing the next work unit, and updating progress. Use when user wants to resume story/task work or mentions /continue command.

Run a Selenium test in visible (non-headless) mode with slowdown so the user can watch it. Use when user wants to demo or visually watch a Selenium test or mentions /demo command.

Preview the planned design for a scenario before writing tests. Shows domain model changes, usecase patterns, and key design choices. Use before red-usecase to get user approval on the implementation approach. If the user rejects the design, offers to discuss inline or escalate to /architecture for a full ADR.

Write a journey summary capturing noteworthy moments (predictions, decisions, surprises, mistakes, quirks). Runs automatically the moment the agent observes something worth noting during work, and manually before /clear or /compact as a final sweep. Idempotent — skips entries already recorded. Use when the agent observes a noteworthy moment, before clearing context, or when the user mentions /handoff.

Re-explain the last thing in plain words. Use ONLY when the user explicitly types /plain or asks to "explain that simply / in plain words / like I'm not an expert". Never invoke on your own initiative — this is a manual button the user presses when something was hard to follow.

Imagine the incident a just-shipped work unit would cause and surface unguarded gaps as follow-ups. Use after a work-unit commit, or when the user runs /premortem.

Scan a prompt file for layer violations and structural problems, then fix them. Use when user wants to clean up documentation files or mentions /prompt-refactor command.

Classify new content (rule, pattern, workflow step) and write it to the correct documentation layer. Use when user wants to add prompt documentation or mentions /prompt-update command.

Execute a QA task's manual checklist against an external environment (prod-copy) by driving a watched, headed browser one action at a time. Use when running a smoke/regression checklist against prod-copy, validating the critical path against real integrations, or when the user mentions /qa-run. Navigation is UI-only — reach every page by clicking, never by typing a direct in-app URL.

Evidence-based root cause analysis for a defect — re-verify every prior assumption, test competing hypotheses, and confirm the cause with real data (logs, test runs, measurements) before any fix. Use for the bug-task `root cause analysis` step, or whenever a cause is claimed but not yet proven.

Refactor code using Martin Fowler's patterns. Improves readability, moves behavior closer to data, removes unnecessary abstractions.

Generate story specifications from MVP stories list. Use when user wants to create detailed story documentation or mentions /story command.

Create a new task (bug, refactoring, or qa) with spec and progress tracking. Use when user wants to create a task or mentions /task command.

Run acceptance tests (backend API, frontend UI, or load suite). Use when user wants to run E2E acceptance tests or mentions /test-acceptance command.

Review tests to replace loose validation (contains, isNotNull, isNotEmpty) with strict validation (isEqualTo on parsed fields). Use when user wants to improve test assertions or mentions /test-review command.

Generate BDD test specifications for story in 6 categories (API, UI, Load, Infrastructure, Security, Integration). Use when user wants to create test cases or mentions /test-spec command.

Align frontend component styling to match HTML mockup pixel-for-pixel. Reads mockup CSS, compares with component, fixes all differences. Use when user wants to match a component to its mockup or mentions /align-design command.

Generate OpenAPI specifications for story endpoints. Use when user wants to create API specs for a story or mentions /api-spec command.

Facilitate architectural decisions during implementation. Use when the developer encounters a design choice with multiple viable approaches (data model, race condition handling, processing strategy, schema design) and needs structured discussion, trade-off analysis, and documentation. Triggers on /architecture or when a conversation reveals competing design options that need evaluation before coding continues.

Kill orphaned chromedriver and headless Chrome processes left over from failed Selenium test runs. Use when Selenium tests mass-fail with Connection reset or TimeoutException, when the system feels sluggish during test runs, or when the user mentions /cleanup-chrome. Also use proactively before running frontend acceptance tests if previous runs were interrupted or stopped.

Review frontend component for hardcoded mockup placeholder data after align-design. Flags emails, dates, prices, and other user-specific values that should be dynamic. Use when user mentions /design-review command.

Document findings, research, or knowledge into ProductSpecification/ with proper cross-references. Use when the user wants to write up research results, create a reference doc, consolidate scattered knowledge, or mentions /doc command. Also triggers on "document this", "write this up", "let's capture what we learned", "create a reference for".

Interactive interview to create interview.md for a story. Asks structured questions about scope, APIs, decisions, and constraints, then generates the handwritten-style context file. Use when user wants to create story specifics or mentions /interview command.

Generate HTML interface mockups for user stories. Use when user wants to create UI mockups, prototypes, or mentions /mockups command.

Run the backend application. Use when user wants to start the backend server or mentions /run-backend command.

Start the frontend dev server. Use when user wants to start the frontend or mentions /run-frontend command.

Take screenshots of HTML mockups from any version. Use when user wants to capture mockups as PNG images, regenerate screenshots, or mentions /screenshot command.

Stop the running backend application. Use when user wants to stop the backend server or mentions /stop-backend command.

Run adapter module tests. First argument is adapter name (matches directory under backend/adapters/). Use when user wants to run adapter tests or mentions /test-adapter command.

Run all tests - unit tests in parallel, then acceptance tests with backend. Use when user wants to run the full test suite or mentions /test-all command.

Run tests with coverage and report uncovered lines/branches per class. Use when user wants to check test coverage, find untested code, identify coverage gaps, or mentions /test-coverage command. Also use after green phases to verify new code is well-covered.

Run frontend tests. Use when user wants to run frontend unit tests or mentions /test-frontend command.

Run use-case module tests quickly. Use when user wants to run use-case tests or mentions /test-usecase command.