productfeeling

name	productfeeling
description	Emotion-aware product design: audit how a product, flow, or surface feels; map emotional journeys; calibrate tone, trust, delight, and friction; write emotion briefs for Impeccable; hand off to DocSlime. Use when the user wants feeling-first product design, emotional UX review, tone/microcopy feel, peak-end moments, trust and anxiety reduction, empty/error/success state emotion, persona emotional lenses, or `/productfeeling` commands. Complements Impeccable (craft) and DocSlime (docs) — does not replace them. Works in chat or with project `docs/` (DocSlime) and optional `.productfeeling/` persistence.
version	0.1.0
argument-hint	[command] [target]
user-invocable	true
allowed-tools	["Bash(node scripts/*)"]
license	AGPL-3.0-or-later

Emotion-aware product design: decide what a product should feel like, then stress-test surfaces, flows, and copy against that north star.

ProductFeeling is the skill. It sits upstream of Impeccable (visual/interaction craft) and beside DocSlime (product docs). Feeling brief → craft → ship.

Catalog: reference/catalog.md

Setup

You MUST do these steps before proceeding:

1. Run node scripts/context.mjs once per session (or node <skill-base-dir>/scripts/context.mjs if the runtime shows the skill path). Keep cwd at the user's project. If the request names a specific surface, flow, or file, append --target <path> to the command. If you've already seen its output in this conversation, do not re-run it. The script prefers a project docs/ tree (PRODUCT, DESIGN, REQUIREMENTS, strategy/, experience/) as the feeling/product source of truth. Legacy FEELING.md is optional. If it reports NO_PROJECT_DOCS: divert into reference/init.md only when the user invoked init, or when they are starting a new feeling definition with no prior context. For scoped commands against material the user already provided in chat (a screen, flow, copy deck, PRD), do not block on init — the user's input is the context. Offer /productfeeling init once as an optional follow-up to persist context into docs. Missing docs or FEELING.md must never block a scoped review.
2. When docs/ exists, load selectively. For ProductFeeling’s own product work (this repo and similar DocSlime trees), prioritize docs/strategy/ then docs/experience/, then PRODUCT/DESIGN — read only the files needed for the current ask. Do not ingest the whole docs tree or the whole handbook by default.
3. If the user invoked a sub-command (audit, map, brief, ...), you MUST read reference/<command>.md next. Non-optional. The reference defines the command's flow.
4. For any review targeting a UI, flow, or codebase artifact, read the actual source material before analyzing. Do not critique from memory or summary alone.
5. Read reference/principles.md when performing any evaluative or craft-facing command. It encodes the emotion principles. Skipping it produces vibes-without-structure.
6. When loading handbook pages (strategies, TTPs, discovery, concepts), follow How to load handbook pages in reference/sequence.md: resolve paths under the skill install directory (<skill-base>/handbook/…, next to this SKILL.md), read teaching prose (What it is / Why it works, or concept Definition / Why it matters) before Do/Don’t, metrics, or Agent skill. Do not treat handbook cards as checklists alone. Do not expect handbook/ at the user’s project root.

Core principles (always apply)

• Feeling is a product requirement. "It just feels off" is a signal to name, not wave away.
• One primary feeling. Secondary tones may support; three+ competing feelings usually means none.
• State every moment. Empty, loading, error, success, and recovery are part of the experience, not afterthoughts.
• Earn delight; buy trust. Delight is sparse and earned. Trust and anxiety reduction are table stakes.
• Never manipulate. Dark patterns, manufactured urgency, and emotional coercion are out of scope — call them out, do not invent them.
• Hand off cleanly. Emotion briefs feed Impeccable for craft and DocSlime for durable product narrative.

Commands

Command	Category	Description	Reference
init	Setup	Capture feeling context into project docs (DocSlime)	reference/init.md
feel [target]	Evaluate	General "how does this feel?" review	reference/feel.md
audit [target]	Evaluate	Structured emotion audit of a surface or flow	reference/audit.md
critique [target]	Evaluate	Scored emotional design critique	reference/critique.md
review [target]	Evaluate	Pre-ship emotional readiness gate	reference/review.md
map [target]	Technique	Emotional journey / valence map	reference/map.md
jobs [target]	Technique	Emotional jobs-to-be-done	reference/jobs.md
tone [target]	Technique	Voice and microcopy emotional tone	reference/tone.md
peaks-ends [target]	Technique	Peak-end rule for key moments	reference/peaks-ends.md
friction [target]	Technique	Emotional friction inventory	reference/friction.md
delight [target]	Technique	Where intentional delight belongs	reference/delight.md
trust [target]	Technique	Trust signals and anxiety reducers	reference/trust.md
states [target]	Technique	Empty / loading / error / success emotion	reference/states.md
persona [target]	Technique	Emotional lens per persona or segment	reference/persona.md
intensity [target]	Technique	Calibrate emotional volume	reference/intensity.md
anti-patterns [target]	Technique	Emotional anti-patterns and dark patterns	reference/anti-patterns.md
metaphors [target]	Technique	Brand emotion metaphors that drive craft	reference/metaphors.md
brief [target]	Craft	Write an emotion brief for Impeccable	reference/brief.md
handoff [target]	Craft	Hand off to Impeccable and/or DocSlime	reference/handoff.md
sequence [target]	Craft	Handbook strategy playbooks (and cross-cutting chains)	reference/sequence.md
next [target]	Meta	Do the next best-practice improvement and keep going, no re-asking	reference/next.md
random [target]	Meta	Roll a random technique to break tunnel vision	reference/random.md
library	Meta	Full "everything I could do here" index (commands + handbook + companions)	reference/library.md

Reference-only: principles.md · catalog.md

Plus management commands: pin <command> and unpin <command>.

Routing rules

1. No argument: an opinionated, non-executing nudge — not the full index (that's library) and not auto-execution (that's next).
   • If NO_PROJECT_DOCS and no usable in-chat context: lead with /productfeeling init as the single best next move, with one line on why.
   • Otherwise: name the single best next move for what the user is working on — use the same judgment next uses to pick it, but stop and describe it, then wait for a go-ahead instead of running it. Give a one-line why.
   • Then offer a short, inviting slate of alternatives: check your work with audit (mechanical completeness) or critique (push for greatness), or try random if you're feeling lucky.
   • Close with a pointer to library for the full "everything I could do" index. Never auto-run a command.
2. First word matches a command (table above or pin / unpin): load its reference and follow it. Everything after the command name is the target.
3. First word doesn't match, but intent maps clearly (e.g. "how does this feel" → feel, "emotion audit" → audit, "make it great / be critical" → critique, "surprise me" → random, "what can you do / list everything" → library, "just keep improving it" → next, "journey map" → map, "write a brief for Impeccable" → brief, "ship checklist" → review): load that command's reference and proceed.
4. No clear match: general ProductFeeling invocation. Apply setup, principles, and the user's full argument as context. Default to feel flow unless a specific technique fits better — ask once if two techniques are equally plausible.

next is the one command that keeps acting without re-asking — it carries standing consent to do the next best-practice piece and continue (within the guardrails in its reference). All other commands still stop at findings unless the user asked for changes.

If the first word is init, or routing rule 3 maps to a from-scratch feeling capture, reference/init.md owns the rest of the flow after setup.

Pin / Unpin

Pin creates a standalone shortcut so $<command> invokes $productfeeling <command> directly. Unpin removes it.

node scripts/pin.mjs <pin|unpin> <command>

Valid <command> is any command from the table above. Report the script's result concisely.

Output conventions

When writing artifacts to .productfeeling/:

• Reviews → .productfeeling/reviews/<slug>-<command>.md
• Session notes → .productfeeling/sessions/<date>-<slug>.md
• Emotion briefs → .productfeeling/briefs/<slug>-brief.md (create briefs/ if needed)
• Use ISO dates in filenames when dating sessions. Slug from the target topic (lowercase, hyphenated).

Always tell the user what was written and where. In chat-only mode (no .productfeeling/ and no docs/), deliver the full analysis in the response; offer to persist with /productfeeling init (into docs when appropriate) if useful.

Docs writes: Prefer durable updates in docs/strategy/, docs/experience/, PRODUCT/DESIGN, or ADRs when the user asked for lasting product/strategy changes. Do not duplicate product truth into .productfeeling/. Do not dump routine reviews into docs/.

Companions

Skill	What it is	Handoff cue
Impeccable (impeccable.style)	Design guidance for AI coding agents — frontend craft, live browser iteration, deterministic detector rules	After brief or handoff
DocSlime (DecisionNerd/DocSlime)	CLI + skill pack to create, fill, review, and publish an opinionated docs/ tree	After handoff when narrative must live in docs
RedTeam (CurateLabs/RedTeam)	Applied critical thinking — premortems, assumptions checks, devil's advocacy, decision reviews (not security testing)	When a feeling decision or north star needs adversarial stress-testing before commitment

Do not re-implement Impeccable polish, DocSlime document fill, or RedTeam critical-thinking facilitation inside this skill. Produce feeling decisions; hand off execution.

What this is NOT

• Not a general frontend design skill (use Impeccable).
• Not clinical psychology, persuasion engineering, or dark-pattern invention.
• Not a license to prioritize "delight metrics" over clarity, accessibility, or user agency.