Scenario Inspiration
Overview
Asked for inspiration, an agent averages. It returns the most typical answer for the brief, so every
user with a similar brief gets the same look. This skill trades that for four phases: prime,
widen, choose, board. Connection and scope: see the scenario skill in this repo. If a
sibling skill named here is missing from your available skills, load it with skill_get (Scenario
MCP) or ask the user to install it (https://github.com/scenario-labs/skills#install); unattended,
proceed from tool schemas and flag the gap.
Four moves carry most of the value. Name the reflex answer for this brief and the answer of someone
merely avoiding that reflex, then rule out both, because an agent told to be original lands on the
second-obvious answer. Randomize the operator, never the subject: what varies is the transformation
applied to the brief, and a random subject is just noise. When you generate candidates yourself
instead of searching for them, ask yourself for a distribution rather than an answer, and work from
its tail. Then make the user pick, because options that never resolve are a gallery, not direction.
Quick reference
| Phase |
Do |
Detail |
| 1. Prime |
Locked constraints, the open axis, both reflexes, what counts as a hit |
references/widen.md |
| 2. Widen |
Four lanes plus wildcard draws; over-gather, near and far |
widen.md, references/sources.md |
| 3. Choose |
Three or four mutually exclusive directions one named axis apart, plus an escape hatch |
references/choose.md |
| 4. Board |
The pick becomes a collection, every reference annotated with its job |
references/moodboard.md |
Four lanes feed phase 2. Ask the user first, for three things they already love and one they cannot
stand; unattended, take those from the task instructions and continue without them if it names none.
Then search with target: "assets" twice, once over the team's own work (public omitted) and
once over the public catalog (public: true), and last the open web, which is not a search call.
Public asset hits carried metadata.prompt, the wording that produced them, and target: "models"
hits carried exampleAssetIds at authoring time: free style research either way.
images: {like, unlike} steers by example in both directions.
Draw far domains yourself with scripts/wildcard.py, run from the skill
directory (python3 scripts/wildcard.py --count 4). A model asked for something random samples its
own habits; the script samples a corpus and prints its seed, so a draw can be replayed or
deliberately never repeated.
Worked example: a puzzle game's world map, nothing decided
- Prime. Locked: 16:9, readable at phone size. Open: everything else. Reflex: candy-colored
isometric islands. Second reflex: the same islands, muted and "cozy". Both are out of bounds now.
A hit is a map whose regions read apart in grayscale. Run the obvious query once and keep it as
the baseline to beat.
- Widen.
search the project for anything already on brand, then public: true at
query_semantic_ratio: 0.8 and again at the keyword default, reading the prompts on the best
hits. Draw four wildcards, run each as its own query, and keep near, middle, and far finds rather
than only the strangest.
- Choose. Name the axis first (here, how the world is depicted), then four positions on it.
Each gets a title, an intent line, two or three references shown with
asset_display, and who it
wins for and when it fails. Say what you cut. Ask which is closest and what they would take from
another, and offer the escape hatch. When nobody can answer, take the pick from the task
instructions, else choose the option that best satisfies the hit line from prime, say which and
why, mark it provisional, and keep going.
- Board.
collection_create the winner, then copy in every foreign reference: public assets
belong to their own team, and collection_add_assets on one returned 403 at authoring time.
asset_download, curl -L, upload_asset. asset_update each with its reason, tag the
anti-references, hand the collection id to scenario-image or scenario-consistency.
Common mistakes
- Skipping prime and searching the brief's own words, which returns the category reflex sorted by
relevance.
- Four options that are one idea in four colors. If picking A does not rule out B, they are not
directions.
- Presenting only the survivors. Naming what was cut, and why, is what makes the rest credible.
- Calling a set varied without checking it against what the literal query already returned.
- Treating every strange find as a keeper. It earns a slot only when you can name what it connects
to and what it is worth.
- Boarding whatever came back. Every reference needs a job (light, color, composition, material,
shape, subject, environment) or it is decoration.
- Mixing lighting worlds on one board: a hard-flash reference and a soft-window reference cancel at
generation time.
- Handing a model an 18-image board. Reduce to three to six role-tagged references first.
- Shipping the undirected default when the brief asked for none of it: subject centered and facing
the lens, posed, evenly lit, everything in focus, a processed gloss. Every axis a prompt leaves
open (framing, moment, lens, time of day, light) is filled with the model's most average answer,
so name each one.
1---2name: scenario-inspiration3description: "Use when a creative task has no direction yet: finding inspiration, exploring looks or styles, proposing concept directions to pick between, building a moodboard or reference board, doing visual research, mining existing assets for a style, or escaping generic AI-looking output. Also when a saved Scenario collection should drive a new batch as art direction. Keywords: inspiration, moodboard, references, art direction, style exploration, concepts, serendipity, ideation, surprise me."4license: MIT5metadata:6 source: https://skillmd.com/skills/scenario-labs/scenario-inspiration7---89# Scenario Inspiration1011## Overview1213Asked for inspiration, an agent averages. It returns the most typical answer for the brief, so every14user with a similar brief gets the same look. This skill trades that for four phases: **prime**,15**widen**, **choose**, **board**. Connection and scope: see the `scenario` skill in this repo. If a16sibling skill named here is missing from your available skills, load it with `skill_get` (Scenario17MCP) or ask the user to install it (https://github.com/scenario-labs/skills#install); unattended,18proceed from tool schemas and flag the gap.1920Four moves carry most of the value. Name the reflex answer for this brief and the answer of someone21merely avoiding that reflex, then rule out both, because an agent told to be original lands on the22second-obvious answer. Randomize the operator, never the subject: what varies is the transformation23applied to the brief, and a random subject is just noise. When you generate candidates yourself24instead of searching for them, ask yourself for a distribution rather than an answer, and work from25its tail. Then make the user pick, because options that never resolve are a gallery, not direction.2627## Quick reference2829| Phase | Do | Detail |30| --------- | -------------------------------------------------------------------------------------- | -------------------------------------------------------- |31| 1. Prime | Locked constraints, the open axis, both reflexes, what counts as a hit | [references/widen.md](references/widen.md) |32| 2. Widen | Four lanes plus wildcard draws; over-gather, near and far | widen.md, [references/sources.md](references/sources.md) |33| 3. Choose | Three or four mutually exclusive directions one named axis apart, plus an escape hatch | [references/choose.md](references/choose.md) |34| 4. Board | The pick becomes a collection, every reference annotated with its job | [references/moodboard.md](references/moodboard.md) |3536Four lanes feed phase 2. Ask the user first, for three things they already love and one they cannot37stand; unattended, take those from the task instructions and continue without them if it names none.38Then `search` with `target: "assets"` twice, once over the team's own work (`public` omitted) and39once over the public catalog (`public: true`), and last the open web, which is not a `search` call.40Public asset hits carried `metadata.prompt`, the wording that produced them, and `target: "models"`41hits carried `exampleAssetIds` at authoring time: free style research either way.42`images: {like, unlike}` steers by example in both directions.4344Draw far domains yourself with [scripts/wildcard.py](scripts/wildcard.py), run from the skill45directory (`python3 scripts/wildcard.py --count 4`). A model asked for something random samples its46own habits; the script samples a corpus and prints its seed, so a draw can be replayed or47deliberately never repeated.4849## Worked example: a puzzle game's world map, nothing decided50511. **Prime.** Locked: 16:9, readable at phone size. Open: everything else. Reflex: candy-colored52 isometric islands. Second reflex: the same islands, muted and "cozy". Both are out of bounds now.53 A hit is a map whose regions read apart in grayscale. Run the obvious query once and keep it as54 the baseline to beat.552. **Widen.** `search` the project for anything already on brand, then `public: true` at56 `query_semantic_ratio: 0.8` and again at the keyword default, reading the prompts on the best57 hits. Draw four wildcards, run each as its own query, and keep near, middle, and far finds rather58 than only the strangest.593. **Choose.** Name the axis first (here, how the world is depicted), then four positions on it.60 Each gets a title, an intent line, two or three references shown with `asset_display`, and who it61 wins for and when it fails. Say what you cut. Ask which is closest and what they would take from62 another, and offer the escape hatch. When nobody can answer, take the pick from the task63 instructions, else choose the option that best satisfies the hit line from prime, say which and64 why, mark it provisional, and keep going.654. **Board.** `collection_create` the winner, then copy in every foreign reference: public assets66 belong to their own team, and `collection_add_assets` on one returned 403 at authoring time.67 `asset_download`, `curl -L`, `upload_asset`. `asset_update` each with its reason, tag the68 anti-references, hand the collection id to `scenario-image` or `scenario-consistency`.6970## Common mistakes7172- Skipping prime and searching the brief's own words, which returns the category reflex sorted by73 relevance.74- Four options that are one idea in four colors. If picking A does not rule out B, they are not75 directions.76- Presenting only the survivors. Naming what was cut, and why, is what makes the rest credible.77- Calling a set varied without checking it against what the literal query already returned.78- Treating every strange find as a keeper. It earns a slot only when you can name what it connects79 to and what it is worth.80- Boarding whatever came back. Every reference needs a job (light, color, composition, material,81 shape, subject, environment) or it is decoration.82- Mixing lighting worlds on one board: a hard-flash reference and a soft-window reference cancel at83 generation time.84- Handing a model an 18-image board. Reduce to three to six role-tagged references first.85- Shipping the undirected default when the brief asked for none of it: subject centered and facing86 the lens, posed, evenly lit, everything in focus, a processed gloss. Every axis a prompt leaves87 open (framing, moment, lens, time of day, light) is filled with the model's most average answer,88 so name each one.