NONoise

Use ONLY when the user explicitly asks Claude to delegate, hand over, pass, or send work to Codex CLI - phrases like "delega a codex", "usa codex", "handover a codex", "passa a codex", "ask codex", "@codex", "/codex", "via codex", or "with codex". Runs `codex exec` non-interactively and returns Codex's final answer verbatim. NEVER invoke this skill unless the user explicitly names Codex or Codex CLI as the delegate.

Use ONLY when the user explicitly asks to delegate a task to GitHub Copilot CLI — phrases like "delega a copilot", "usa copilot", "ask copilot", "passa a copilot", "fai con copilot", "@copilot", "/copilot", "via copilot", "with copilot". Runs `copilot -p` non-interactively, defaulting to Claude Opus 4.7, so the token cost of large/expensive work (big-file reads, broad explanations, shell-output analysis) bills against the user's GitHub Copilot subscription instead of their Claude subscription. Acts as a transparent passthrough: build a full prompt with the needed context, execute `copilot`, return the response verbatim inside a fenced block. NEVER invoke this skill on your own — only when the user explicitly names Copilot or uses the trigger phrases above. If the user just asks to read a big file or run a long command without naming Copilot, do NOT trigger this skill.

Use ONLY when the user explicitly asks Claude to delegate, hand over, pass, or send work to Gemini CLI — phrases like "delega a gemini", "usa gemini", "passa a gemini", "fai con gemini", "ask gemini", "@gemini", "/gemini", "via gemini", "with gemini". Runs `gemini -p` non-interactively and returns Gemini's answer verbatim. The point is to offload token-heavy work (long file analysis, broad codebase Q&A, big log summaries) to the user's Gemini account quota instead of the Claude subscription. NEVER invoke this skill unless the user explicitly names Gemini or Gemini CLI as the delegate.

Use ONLY when the user explicitly asks the AI assistant to route, dispatch, fan-out, split, or parallelize a task across multiple AI model CLIs (Claude / Codex / Gemini / Copilot's many models) — phrases like "usa lo swarm", "swarm:", "/swarm", "instrada", "fan-out", "route this task", "scegli il modello migliore per", "delega allo swarm", "best model for this", "ensemble", "second opinion da", "team mode", "parallel-dev", "sviluppa in parallelo", "swarm parallel", "fai team", "build with a team". Four modes: single dispatch, sequential pipeline, parallel fan-out, and parallel-team. Works across any harness that can read this SKILL.md (Claude Code, Copilot, Gemini CLI, Codex) — the dispatch mechanism differs per harness (see Harness-specific dispatch). Hierarchical orchestration only — there is NO peer-to-peer messaging between different model CLIs. NEVER invoke this skill on your own; only when the user explicitly uses one of the trigger phrases above.

any input (code, docs, papers, images, videos) to knowledge graph. Use when user asks any question about a codebase, documents, or project content - especially if graphify-out/ exists, treat the question as a /graphify query.

any input (code, docs, papers, images, videos) to knowledge graph. Use when user asks any question about a codebase, documents, or project content - especially if graphify-out/ exists, treat the question as a /graphify query.

Produces and incrementally maintains a versioned reverse-engineering dossier for a subject — a legacy codebase, a microservice, a third-party API, an internal library, a data pipeline. Output is markdown chapters under `docs/support/reverse/<subject-slug>/`, combining project documents (sprint inputs, meeting notes, functional specs) with the subject's code/artifacts indexed via the graphify knowledge graph. Flow — (1) graphify indexing loads context only, (2) long-running interactive Q&A on graph and source, (3) explicit save trigger ("save", "ok write it", "make v1.5") bumps minor version, updates only chapters in the consolidated intent, appends a changelog entry. Triggers — "reverse-engineer X", "document legacy Y", "build a dossier on vendor Z's API", "refresh the reverse on <subject>", "new version of the reverse", or mention of a path under `docs/support/reverse/` in an update context. Do NOT use for one-shot graph queries or single-chapter manual edits.

Acceptance Test Runner — reads acceptance criteria from a sprint manifest and/or requirements files, generates a machine-readable testbook, executes it via Playwright against the running app, produces markdown reports with screenshots, and maintains a backlog of failures. Optionally pushes failures to an external tracker (Paseo integration planned as future work). Use this skill when the user mentions acceptance testing, testbook, running tests against the app, producing a stakeholder-friendly test plan, verifying the app against the sprint manifest, or pushing test failures to a tracker. Triggers — `/atr generate`, `/atr run`, `/atr doc`, `/atr push`, "run acceptance tests", "generate testbook for sprint N", "check the app against the manifest", "create a readable test plan for stakeholders". Language — reports and testbook descriptions follow the project's working language (inferred from the sprint manifest / requirements content, or from `nonoise.config.json` `language` if set; defaults to English).

NONoise SDLC advisor — reads the project-local .nonoise/sdlc-flow.md, tells the user where they are in the flow and what skill to engage next, with a copy-pasteable example prompt. Use when the user types `/polly`, says "start polly" / "avvia polly" / "run polly", or asks an ambiguous "where do I start?" / "what's next?" / "sono perso" and the intent does not map cleanly to a single skill. Polly is a one-shot advisor — produces one message then terminates. Does not orchestrate across turns, does not persist state, does not auto-trigger. The default path for any clear user intent is to engage the matching skill directly; Polly is a fallback for ambiguous cases.

Dev tool for the NONoise maintainer — audits the default `.nonoise/sdlc-flow.md` template against the actually-bundled skills and reports drift. Use when you add or remove a skill in `packages/skills/` or change `MVP_SKILL_BUNDLE` in `packages/create-nonoise/src/scaffold.ts`, to decide if the flow template needs a new phase, a deleted phase, or a renamed `skill` field. Also use when the user says "check sdlc-flow", "audit polly flow", "sync flow template", or after syncing a vendored pack via `node scripts/sync-vendor.mjs`. Scoped to THIS monorepo only — never invoke in a scaffolded project. Proposes a diff against the template; human applies it.

Generate and keep in sync the three meta-docs of a NONoise project — `AGENTS.md` (root, cross-tool, source of truth), `CLAUDE.md` (root, Claude Code adapter), and `.github/copilot-instructions.md` (Copilot adapter). Treats `AGENTS.md` as the single source of truth and preserves skill-owned managed blocks verbatim. Use whenever the user says "generate CLAUDE.md", "generate AGENTS.md", "generate copilot-instructions.md", "sync the meta-docs", "the docs are out of date", "AGENTS.md changed — regenerate the others", "bootstrap docs from scratch", "import this project into NONoise conventions", "reconcile CLAUDE.md and AGENTS.md", or when they start from an existing codebase and want NONoise-style meta-docs produced consistently. Trigger proactively when the user edits `AGENTS.md` and asks for "the others to match", or when the three files visibly disagree.

Dialogic architectural brainstorming skill. Helps an architect study how to implement something starting from just an area name, exploring the existing codebase, reading the project's source of truth at `docs/architecture/` plus `docs/requirements/` and `docs/calls/`, and producing a narrative PRD (story of the decision, inline Mermaid diagrams, explicit anti-scope) under `docs/prd/<area>/`. USE AS STEP 1 of the end-to-end architectural workflow — brainstorm → arch-decision (Quint FPF validation via `quint-fpf`) → sprint-manifest (promotion to sprint). Triggers — "I want to study how to implement X", "how should we handle Y", "brainstorm on area Z", "open a PRD for W", "arch-brainstorm area <slug>". Also triggers without explicit mention when the user is defining a new architectural work area from scratch and wants to think it through together.

Step 2 of the NONoise architectural workflow — validates a `draft` PRD (from `arch-brainstorm`) using Quint FPF via the `quint-fpf` sub-skill. Extracts hypotheses, computes R_eff, writes a per-phase audit folder at `docs/prd/<area>/audit/NN-<study>-fpf/`, presents a mandatory Phase 5.5 human review/approve gate (with override), then updates the PRD frontmatter to `validated` or `rejected`. On PASS, Polly offers `arch-sync` to project the decision into `docs/architecture/` with diff preview — never automatically. Does not create PRDs from scratch (that is `arch-brainstorm`'s job). Triggers — "validate this PRD", "apply Quint FPF to <file>", "arch-decision <path>", "formally assess the architectural decision in <file>", "verify PRD X with FPF". Also triggers without explicit mention when the user refers to a just-written PRD and wants to validate it.

Push the LLM to reconsider, refine, and improve its recent output via structured elicitation methods (Socratic, first principles, pre-mortem, red team, SCAMPER, and 25+ others). Use when user asks for deeper critique, wants to stress-test a draft, or explicitly mentions one of the supported methods. Invoked by the bmad-agent-* personas (Isa, Alex, Daniel, Giulia) to enhance their outputs.

Strategic business analyst and requirements expert. Use when the user asks to talk to Isa or requests the business analyst. Drives requirements elicitation, market/domain/technical research, product brief drafting, and stakeholder PRFAQ exercises.

Dialogic skill that turns raw business input (PDFs, DOCX, markdown notes, stakeholder emails, call transcripts, meeting summaries) into structured requirement files under `docs/requirements/<domain>/<feature>.md`, following NONoise conventions. USE AS PRE-STEP before `arch-brainstorm` — it converts unstructured stakeholder material into a stable set of functional/business requirements the architect can safely reason about. Also produces cross-references to source material parked in `docs/calls/` or `docs/support/`. Triggers — "ingest these requirements", "turn this document into requirement files", "process this call transcript", "extract requirements from <file>", "requirements-ingest", "break this brief into features". Also triggers without explicit mention when the user hands over raw documents (PDF, DOCX, email text) and asks to "organize them into the project" or "prepare for arch-brainstorm".

Step 3 of the NONoise architectural workflow. Promotes `validated` PRDs from `docs/prd/<area>/` to `docs/sprints/Sprint-N/<area>/` and produces one aggregated Sprint Manifest at `docs/sprints/Sprint-N/sprint-manifest.md` listing all promoted PRDs, derived user stories, macro functional tasks (vertical slices) with confidence CL1/CL2/CL3, CL1 tasks flagged for decision records, and cross-area dependencies. Also handles impact analysis when a new PRD is added to an already-open sprint. USE when the user wants to promote validated PRDs to a sprint, talks about "sprint manifest", "sprint plan", "promotion to sprint", "task breakdown macro", "impact analysis on sprint", or "update sprint with new PRD". Triggers — `sprint-manifest area <area> sprint <N>`, `sprint-manifest sprint <N>`, `sprint-manifest sprint <N> update`. Also triggers without explicit mention on "promote <area> to sprint <N>" or "create sprint <N> plan".

Observability-driven debugging, triage, and performance analysis. Ingests traces, logs, and exceptions from any configured backend (App Insights, Datadog, Grafana+Loki, CloudWatch, OpenTelemetry collector, or a generic log file), correlates them by request/trace ID, walks the call chain, identifies the root cause in source code, and proposes a fix with file:line precision. Backend-agnostic via adapter pattern — same methodology across backends. USE whenever the user mentions errors, latency, production incidents, or triage — even without naming a tool. Triggers — pasted correlation/trace/operation/request IDs, browser console errors, HTTP status codes (500, 401, 403, 504, timeout), performance complaints ("is slow", "is lagging"), phrases like "what's broken", "why isn't this working", "check the logs", "triage this", "find the exception for request Y", "what changed in prod in the last hour".

Meta-skill that coaches the AI + user pair through any operational task (deploy, pipeline, provision, migrate, rollback, diagnose, debug-wiring) in any stack, then crystallizes the successful workflow into a project-local reusable skill. Asks WHAT the user wants to achieve, proposes an access menu (CLI > API > Web), gathers context from `docs/architecture/` + optional graphify index, proposes a dry-run plan with risk annotations, executes in paired or delegated mode, and — on success — invokes `skill-creator` to save the command sequence as a project-local skill under `.claude/skills/<slug>/SKILL.md`. Triggers — "deploy this service", "create a CI/CD pipeline for X", "set up rollback", "provision dev env", "run this migration", "automate this ops thing I do manually", "ops on Azure / AWS / GCP / kubernetes", or turning a manual op into a repeatable team skill. Also triggers without explicit mention when the user describes a repeated manual ops workflow with no fitting deploy skill available.

Generic projector of a validated PRD's "Impact on docs/architecture/" checklist into the actual files. Reads the validated PRD plus its `arch-decision` audit folder (specifically `05-decision.md`), parses the strict `[file: NN.md]` format from the appended `## Impact on docs/architecture/` section, proposes a unified diff per target file, applies only the diffs the architect approves, and writes a sync report under `docs/architecture/sync-reports/`. Stack-neutral — knows no target constraints, never validates anything, never reads source code. Triggers — invoked by Polly after `arch-decision` PASS (the architect chooses "Sì — invoco arch-sync" at the post-finalize menu); also invokable manually with phrases like "arch-sync <prd-path>", "project the validated PRD into docs/architecture/", "sync architecture from <prd-path>", "recepisci il PRD validato <path>".

Quint First Principles Framework — structured reasoning methodology with 6 phases (Initialize → Abduct → Deduce → Induce → Audit → Decide), R_eff via WLNK, and Congruence Level tracking. Use when you need to formally validate an architectural decision, a PRD, or any non-trivial claim. Invokable standalone or delegated from other skills (e.g. `arch-decision`). Every phase emits a dedicated markdown file under a single per-cycle folder so the entire reasoning trail is human-readable, git-diffable, and removable in one `rm -rf`. Has two execution modes — tooled (quint MCP server enforces holon state in `.quint/`) and conversational (markdown-only) — but the document trail is produced in both.

System architect and technical design leader. Use when the user asks to talk to Alex or requests the architect. Drives architecture brainstorming, formal validation (Quint FPF), source-of-truth documentation, implementation readiness checks, and LSP advice.

End-to-end release flow for the `create-nonoise` CLI — creates a changeset, runs `pnpm run version` to bump `packages/create-nonoise/package.json` + CHANGELOG, updates the GitHub Release tarball URL in both READMEs, commits + pushes to `main` (which triggers `auto-tag.yml` → `release.yml` → npm publish with provenance), builds and packs the tarball locally, creates the matching GitHub Release with the `.tgz` attached, then polls the workflow until complete. Use whenever the user says "bump create-nonoise", "rilascia nuova versione", "pubblica", "new version", "release 0.X.Y", "bump and publish", "fai il bump", or asks to cut a new version of the CLI. Scoped to THIS monorepo only — do not invoke in scaffolded projects. Also handles the npm-unpublish 24h lock failure mode by surfacing the error and suggesting `gh run rerun --failed`.

Translates macro functional tasks from a NONoise sprint manifest (`docs/sprints/Sprint-N/sprint-manifest.md`) into work items on an external issue tracker via a pluggable adapter pattern. Ships adapters for GitHub Issues (default), Azure DevOps, Jira, Linear, plus `dry-run` mode that prints what would be created without touching any remote system. Idempotent — searches for existing work items matching a task ID before creating new ones. Use when the user mentions "export to tracker", "create issues from sprint manifest", "push work items", "sync sprint to GitHub / Azure DevOps / Jira / Linear", "spec to workitem", "dry-run work items", or wants to translate a manifest into an external backlog. Commands — `/spec-to-workitem sprint N`, `/spec-to-workitem sprint N adapter=github-issues`, `/spec-to-workitem sprint N dry-run`. Auth is always per-adapter via environment variables — never hardcoded.

Produces and maintains living C4 architecture diagrams (Context / Container / Component / Code) for a NONoise project via a single Structurizr DSL workspace at `docs/architecture/c4/workspace.dsl`, from which all views are regenerated. Source-controlled, diff-able — unlike static Mermaid. USE right after `arch-decision` returns PASS on a PRD so diagrams stay in sync with the validated decision. Triggers on "/c4 update", "update c4", "update C4 diagrams", "generate C4 diagrams", "refresh the architecture diagrams", "structurizr", "rebuild the container view", "sync the C4 with the new PRD". Also triggers without explicit mention when the user wants to refresh architectural diagrams after a PRD/ADR was validated. Stack-neutral — works for any project with populated `docs/architecture/` (Angular+dotnet, Node+React, Python, Go, Flutter, data pipelines, CLIs).

Formal validation of IT functional requirements via a curated suite of 10 methodologies (INVEST, SMART, EARS, MoSCoW, 5W1H, WOTIF, IEEE 830/29148, FMEA-Lite, SWOT, ATAM-Lite). Called by Isa (bmad-agent-analyst) as capability RV — NOT self-activating. Classifies the requirement, recommends a 2–5 methodology subset, waits for analyst confirmation, runs only the selected ones, deduplicates gaps, and generates business-friendly challenge questions. Skippable in quick flows.

Technical documentation specialist and knowledge curator. Use when the user asks to talk to Daniel or requests the tech writer. Transforms complex concepts into accessible documentation — READMEs, user guides, API docs, Mermaid diagrams, plan documents, and brownfield project documentation.

UX designer and UI specialist. Use when the user asks to talk to Giulia or requests the UX designer. Drives UX planning, interaction design, design-system authoring (DESIGN.md), UI implementation handoff, critique, and polish.

Generates a DESIGN.md design-system document for the current project following the Stitch DESIGN.md format popularized by getdesign.md (VoltAgent/awesome-design-md). Use when the user asks to "create a design.md", "generate design system docs", "write DESIGN.md", "bootstrap UI style guide", "document the design system for AI agents", or when starting a frontend project and no design spec exists yet. Can start from scratch via guided Q&A or clone an existing community DESIGN.md as a starting point.

Discovers AI skills from a curated registry of external sources (Anthropic official, Claude plugin marketplaces, community repos, awesome-lists) and installs them into the current project. Use when the user asks to "find a skill for X", "do we have something for Y?", "add <technology>/stack support", "install a skill from GitHub", "what skills could help with this codebase?", or when Polly needs to plug a gap by fetching a skill that isn't bundled in the framework. Also use proactively after detecting the project stack (Angular, C#, Python, Go, etc.) to suggest stack-specific skills the framework does not ship by default.

Adapt designs to work across different screen sizes, devices, contexts, or platforms. Implements breakpoints, fluid layouts, and touch targets. Use when the user mentions responsive design, mobile layouts, breakpoints, viewport adaptation, or cross-device compatibility.

Review a feature and enhance it with purposeful animations, micro-interactions, and motion effects that improve usability and delight. Use when the user mentions adding animation, transitions, micro-interactions, motion design, hover effects, or making the UI feel more alive.

Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity ratings and actionable plan. Use when the user wants an accessibility check, performance audit, or technical quality review.

Amplify safe or boring designs to make them more visually interesting and stimulating. Increases impact while maintaining usability. Use when the user says the design looks bland, generic, too safe, lacks personality, or wants more visual impact and character.

Improve unclear UX copy, error messages, microcopy, labels, and instructions to make interfaces easier to understand. Use when the user mentions confusing text, unclear labels, bad error messages, hard-to-follow instructions, or wanting better UX writing.

Add strategic color to features that are too monochromatic or lack visual interest, making interfaces more engaging and expressive. Use when the user mentions the design looking gray, dull, lacking warmth, needing more color, or wanting a more vibrant or expressive palette.

Evaluate design from a UX perspective, assessing visual hierarchy, information architecture, emotional resonance, cognitive load, and overall quality with quantitative scoring, persona-based testing, automated anti-pattern detection, and actionable feedback. Use when the user asks to review, critique, evaluate, or give feedback on a design or component.

Add moments of joy, personality, and unexpected touches that make interfaces memorable and enjoyable to use. Elevates functional to delightful. Use when the user asks to add polish, personality, animations, micro-interactions, delight, or make an interface feel fun or memorable.

Strip designs to their essence by removing unnecessary complexity. Great design is simple, powerful, and clean. Use when the user asks to simplify, declutter, reduce noise, remove elements, or make a UI cleaner and more focused.

Make interfaces production-ready: error handling, empty states, onboarding flows, i18n, text overflow, and edge case management. Use when the user asks to harden, make production-ready, handle edge cases, add error states, design empty states, improve onboarding, or fix overflow and i18n issues.