superomni

Turn a fuzzy idea into a concrete, reviewable spec via structured dialogue. THINK stage entry. Triggers: "brainstorm", "design", "spec this out", "let's think through".

Multi-lens auto-review of an IMPLEMENTATION PLAN (not code). Strategy → Design (if UI) → Engineering phases. All decisions auto-resolved via 6 Decision Principles. Triggers: "review plan", "autoplan".

Post-ship documentation update. Cross-references the diff, updates README, ARCHITECTURE, CONTRIBUTING, CLAUDE.md to match what shipped. Polishes CHANGELOG voice, cleans up TODOs, optionally bumps VERSION. Triggers: "update docs", "sync documentation", "post-ship docs", "update the readme", "document what shipped". Proactively suggest after a PR is merged or code is shipped.

Meta-skill: create, install, list, and manage skills and agents within the superomni framework. Merges writing-skills + agent-management into one unified workflow. Triggers: "create skill", "write a skill", "install skill", "list skills", "create agent", "write an agent", "install agent", "list agents", "new skill", "new agent", "add skill", "add agent", "manage framework".

Combined RELEASE stage (ship + retrospective) in a single artifact. Triggers: "/release", "end sprint", "ship and reflect". Replaces the retired ship skill.

Use when turning a spec or idea into a concrete, executable implementation plan. Triggers: "write a plan", "plan this out", "create implementation plan", "how should we implement".

Unified entry point: detect current pipeline stage, launch guided workflow, auto-delegate to next skill. Triggers: "/vibe", "activate framework", "start workflow". NOT a reference document — use workflow for that.

Safety guardrails that activate before high-risk operations. Prevents destructive, irreversible, or security-sensitive changes without explicit confirmation. Triggers: destructive operations, production configs, database migrations, security-sensitive files, "be careful", "careful mode".

Structured self-check against acceptance criteria BEFORE claiming DONE. Triggers: "I'm done", "verify this", "finished". NOT for sprint retrospection — use self-improvement for that.

Use when a feature branch is complete and ready to merge. Performs final cleanup, verification, and merge preparation. Triggers: "finish branch", "ready to merge", "branch complete", "merge this".

Meta-skill: explains the superomni skill framework to the AI agent. Always active. Read this first.

Structured P0/P1/P2 code review of WRITTEN CODE (not plans). Three modes: giving / receiving / security. Triggers: "review this code", "ready for PR", "address review", "security audit". NOT for reviewing plans — use plan-review for that.

Dependency security, license, and freshness audit. Dispatches dependency-auditor agent to scan all package managers. Triggers: "dependency audit", "check dependencies", "npm audit", "security scan", "check for vulnerabilities", "outdated packages", "license check".

Execute a WRITTEN PLAN step-by-step with verification gates and wave-based sub-agent dispatch. BUILD stage orchestrator. Triggers: "execute plan", "run the plan", "start executing". NOT for writing new code from scratch — use test-driven-development for that.

Create distinctive, production-grade frontend interfaces with high design quality. Use when building UI components, pages, or applications. Guides from context gathering through implementation to quality gate. Triggers: "frontend design", "build UI", "design this page", "make it look good"

Use when working on multiple features or branches simultaneously. Git worktrees allow multiple branches to be checked out at the same time. Triggers: "work on multiple features", "parallel branches", "worktree".

Design, audit, and improve the agent harness — the environment, constraints, context management, evaluation gates, and feedback loops that surround and guide AI agents. Inspired by OpenAI and Anthropic harness engineering best practices. Triggers: "harness", "harness audit", "improve harness", "agent environment", "context management", "evaluation gates", "feedback loop", "harness engineering".

Use for open-ended exploration and investigation of unfamiliar codebases, systems, or problems. Complements systematic-debugging (which is for known errors). Use investigate when: "how does X work", "understand this codebase", "map this system". Triggers: "investigate", "explore", "understand how", "map the codebase".

Product discovery and idea validation (startup/builder modes). Outputs a design-doc for product decisions, NOT a sprint spec. Triggers: "office hours", "validate my idea", "help me think through a product".

Use before deploying any code to production. Verifies observability, reliability, operability, and goal alignment. The gate between "tests pass" and "safe to deploy". Triggers: "production ready", "ready to deploy", "pre-deploy check", before using ship skill.

Use when performing quality assurance on code changes. Identifies test gaps, writes missing tests, explores edge cases, and auto-fixes with caution. Triggers: "QA", "quality check", "test coverage", "run QA", "check quality".

Systematic, behavior-preserving code refactoring with safety gates. Dispatches refactoring-agent. Triggers: "refactor", "clean up code", "reduce tech debt", "extract method", "rename". NOT for reactive PR feedback — use code-review for that.

Canonical reflection skill for post-task and post-sprint review. Applies first-principles reflection across process quality and delivery history. Produces improvement reports in docs/superomni/improvements/ and can generate retrospective evidence. Triggers: "self-improve", "evaluate performance", "reflect on execution", "how did we do", "what could be better", "evaluate this sprint", "improve process", "first principles review", "/retro", "weekly retro", "what did we ship", "engineering retrospective".

Capture user's code style preferences through examples, not rules. Generates a style profile for consistent code generation across sessions. Triggers: "/style-capture", "capture style", "show me your preferences", "style profile", project initialization (new project detected by vibe).

Default working mode for non-trivial tasks. Decompose into waves of 5-10 parallel sub-agents. Triggers: any multi-file / multi-concern task, "use subagents", "parallel implementation".

Use when encountering any bug, test failure, or unexpected behavior. Iron Law: no fixes without root cause investigation first. Four phases: Investigate → Analyze → Hypothesize → Implement. Triggers: any error, test failure, "it's broken", "why isn't X working".

Globally enforced Red-Green-Refactor for WRITING NEW CODE. Triggers on: "implement", "add feature", "fix bug", any code-writing step. NOT for task decomposition — use subagent-development for that.

Reference stub pointing to using-skills and vibe auto. The pipeline's operational logic now lives in `vibe` (runtime) and `using-skills` (routing + decision principles). Triggers: "workflow", "pipeline", "what runs at which stage".