agentic-engineering-patterns

Generates or upgrades a project's e2e-test skill to the three-tier BDD layer-gate shape. Use for "scaffold e2e", "set up e2e testing", "add BDD journeys", or thin/bash migration; normally delegated by /aep-scaffold.

Implements one feature autonomously in an isolated workspace, including task commits, review, tests, PR, feedback, and merge. Use when a workspace agent should build or implement; this workflow does not interact with the user.

Designs a feature interactively on the integration branch with OpenSpec exploration, proposal, review, and committed artifacts. Use for "design a feature" or "explore and propose"; next run /aep-launch.

Launches an autonomous workspace agent for a designed feature by creating its worktree, selecting an /aep-executor backend, and optionally adding an evaluator. Use after /aep-design or for "launch workspace" and "spawn agent".

Archives OpenSpec artifacts and cleans the workspace after a PR merges. Use for "wrap up", "archive", or "post-merge cleanup"; this is the final feature step before /aep-dispatch or /aep-reflect.

Runs the dispatch-to-wrap cycle continuously for one layer from the main workspace. Use for "autopilot", "hands-free", or unattended multi-story execution; prefer /aep-dispatch for one story only.

Spawns and steers workspace agents across Claude Code and Codex backends. Used by /aep-launch, /aep-build, and /aep-autopilot; invoke directly for backend choice, launch mode, host/tmux detection, or workflow execution.

Canonical generator/evaluator pattern for artifact scoring, agent contracts, eval protocol, and findings. Used by /aep-build, /aep-validate, and /aep-launch; invoke for "gen/eval", "separate evaluator", or artifact review.

Selects and routes one ready story by syncing signals, scoring the queue, and handing off to /aep-design or /aep-launch. Use for "what's next", "dispatch", or "pick a story"; use /aep-autopilot for continuous execution.

Validates and frames a product opportunity for downstream agents. Use to start or revisit an idea, validate product direction, answer "what should we build", or create the Context Document for /aep-map and /aep-design.

Decomposes a product into a system map, layered story graph, and agent topology. Use after /aep-envision or for "decompose", "story map", "system architecture", "break this down", and "plan the stories".

Builds a human-reviewed OOUX/ORCA object map and noun-first UI blueprint. Use after /aep-map, before UI stories, or for "object model", "object map", "OOUX", "ORCA", and "noun-first IA".

Classifies post-ship feedback and routes it to the story graph, architecture, or opportunity hypothesis. Use after /aep-wrap or user testing, or for "reflect", "what did we learn", "update product context", and "replan".

Monitors bug trackers, errors, and telemetry from the main workspace, then classifies, deduplicates, and creates backlog stories. Use for "watch", "monitor for new work", "ingest errors", or telemetry-driven stories.

Creates a new AEP project or converges an existing project's agentic infrastructure. Use for new project, scaffold, infrastructure setup, or skills-layout repair; use /aep-onboard for installation and orientation.

Canonical AEP git and worktree reference covering base resolution, lifecycle, naming, task commits, PRs, and recovery. Use when another skill needs a git convention or the user asks how AEP uses git.

Reviews product UI with theory-grounded guidelines and a 0–4 heuristic severity check. Use for design, usability, heuristic, or accessibility review; not taste capture (/aep-calibrate) or object IA (/aep-model).

Captures process or stack learnings downstream and routes them back to AEP. Use after /aep-wrap or /aep-reflect, or for "workflow feedback", "capture learnings", and "pull learnings upstream".

Authors a deterministic multi-agent harness for one large, uncertain, or verification-heavy task. Use for "dynamic workflow", "ultracode", or "orchestrate subagents"; not process feedback (/aep-workflow-feedback).

Captures human judgment for a quality dimension agents cannot evaluate. Use when /aep-reflect finds a works-vs-right gap, before a .5 alignment layer, or for "calibrate" and "design brief".

Validates an AEP artifact with separate generator, evaluator, and optional protocol-checker roles. Use for code review, implementation validation, document or RFC validation, after /aep-envision, /aep-map, or /aep-design, or for "validate", "verify", "dry-run", "evaluate", and "gen/eval".

Installs, verifies, and explains Agentic Engineering Patterns (AEP). Use for first-time AEP or plugin mentions, environment setup, tool checks, project plugin configuration, or an overview; use /aep-scaffold to create a project.