art-director
Skill243 starsChanged 3 months agoMITView on GitHub
Orchestrate iterative visual style searches with branch prompts, decision graphs, feedback loops, and final direction selection.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- fralapo/awesome-agent-skillstheir own version, first seen Nov 8, 2026
- NewenergyEric/art-director-skilltheir own version, first seen Oct 17, 2026
- hcaiano/skillstheir own version, first seen Oct 6, 2026
- mcne0144/ClaudeSkillstheir own version, first seen Oct 4, 2026
- serejaris/personal-corp-osword for word, first seen Aug 6, 2026
- ItsssssJack/seven-skillstheir own version, first seen Aug 1, 2026
Other versions7 projects use a version of this.
Word for word · 1
- serejaris/personal-corp-os★ 243100% same
Edited · 0
None found.
Wrote their own version · 6
- ItsssssJack/seven-skills★ 19same name
- NewenergyEric/art-director-skill★ 5same name
- hcaiano/skills★ 3same name
- fralapo/awesome-agent-skills★ 1same name
- lizzyDeng/claude-skills★ 0same name
- mcne0144/ClaudeSkills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository31 other files it gives its agents.
- AGENTS.mdAGENTS.md
- audit-agent-rulesSkill
- avatarSkill
- benchmarkSkill
- cc-analyticsSkill
- ceo-councilSkill
- claude-md-writerSkill
- claude-sessions-syncSkill
- corp-doctorSkill
- corp-newSkill
- dailySkill
- fable-ruki-agentySkill
- gh-issuesSkill
- grill-meSkill
- harnessSkill
- html-draftSkill
- ideaSkill
- it-ram-doctorSkill
- make-3dSkill
- make-landingSkill
- managerSkill
- meeting-copilotSkill
- readme-generatorSkill
- retroSkill
- safe-public-releaseSkill
- sdlc-milestonesSkill
- tg-bot-opsSkill
- to-issuesSkill
- to-prdSkill
- weekly-planningSkill
- weekly-retroSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
25 rules. 0 are common, 16 are distinctive, and 9 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
16
Distinctive
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill when the work is an art-direction search for a media production workspace.
- Very specific: line 39
If the skill was installed without assets/style-search-map-template.html, create the same folders and write a simple local HTML preview …
- Very specific: line 41
If the run continues an existing direction, increment vNN and make every new node point back to its parent decision.
- Distinctive: line 57
Keep the important path explicit: iteration checkpoint -> option nodes -> selected option node -> next iteration checkpoint.
- Very specific: line 61
Read the current project files: index.html, latest vNN-*.html, process/v*/, prompts/v*/, styles/, and relevant research/.
- Very specific: line 64
Write prompts in prompts/vNN/.
- Very specific: line 66
Create or update the HTML direction map as the first section of vNN-<slug>.html.
- Distinctive: line 67
Put preview cards under the map with anchors matching tree node ids.
- Very specific: line 69
Create the next iteration from feedback: feedback -> interpretation -> new branch -> asset -> decision.
- Distinctive: line 74
Every style-search HTML page starts with a direction graph that answers:
- Very specific: line 83
Use skills/art-director/assets/style-search-map-template.html as the starter preview when the host repo has no renderer yet.
- Distinctive: line 109
Use manual SVG layout only as a fallback or baseline.
- Distinctive: line 181
Every iteration cluster shows brief, looked-at variants, options, carry-forward, and dropped motifs.
- Distinctive: line 189
Every node with a representative image opens that image in a detail panel or links to a visible preview card.
- Distinctive: line 191
Avoid duplicating the same graph nodes immediately below as cards or a second table.
- Distinctive: line 195
When feedback arrives after a preview:
- Very specific: line 199
Add or update the iteration cluster's feedback, carry_forward, and drop fields.
- Distinctive: line 201
Create the next iteration cluster with the changed brief and option leaves.
- Very specific: line 253
If the reviewer rejects the whole pass, make a new root child named reset and record the failure mode before generating again.
- Distinctive: line 257
Keep private chats, private screenshots, auth material, billing records, customer records, infrastructure details, and non-public …
- Distinctive: line 258
Use repo-relative paths in run logs and examples.
- Distinctive: line 265
Run the narrow checks that match the pass:
- Distinctive: line 276
If the preview uses an external CDN, also check the pinned runtime:
- Distinctive: line 291
every tree node with an asset links to a visible card.
- Distinctive: line 295
every referenced asset path resolves or is marked as pending.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Art Director
Use this skill when the work is an art-direction search for a media production workspace.
The skill coordinates this loop:
brief -> references -> generation branches -> HTML preview -> feedback -> next branches -> selected directionFor single-image production, use the host repo's image-generation or media-run instructions. For comic pages, albums, storyboards, or platform-specific assets, use the host repo's format-specific playbook when one exists.
Default Output Shape
For a project under art-direction/<project>/, each serious search pass should leave:
art-direction/<project>/
prompts/vNN/YYYY-MM-DD-vNN-<slug>-prompts.md
process/vNN/YYYY-MM-DD-vNN-<slug>-log.md
styles/YYYY-MM-DD-<style-system>.md
assets/vNN-<slug>/
<variant-id>.png
vNN-<slug>.htmlFor a first run in an empty workspace, create the minimal folders before writing prompts:
mkdir -p art-direction/<project>/{prompts/v01,process/v01,styles,assets/v01-style-search}
cp skills/art-director/assets/style-search-map-template.html art-direction/<project>/v01-style-search.htmlIf the skill was installed without assets/style-search-map-template.html, create the same folders and write a simple local HTML preview from the cluster schema in this file.
If the run continues an existing direction, increment vNN and make every new node point back to its parent decision.
The visual map must be an iteration-cluster direction graph:
idea
-> iteration cluster
-> option leaf nodes
-> selected option node
-> carry forward / drop
-> next iteration cluster
-> option leaf nodes
-> selected option node
-> outputKeep the important path explicit: iteration checkpoint -> option nodes -> selected option node -> next iteration checkpoint. Unselected options stay as leaf nodes with their status.
Workflow
- Read the current project files:
index.html, latestvNN-*.html,process/v*/,prompts/v*/,styles/, and relevantresearch/. - Identify the active question: what style, artifact, audience, format, or campaign is being decided.
- Build an iteration cluster before generation: incoming brief, 3-8 option leaves, difference axes, expected selection criteria, and what may carry forward.
- Write prompts in
prompts/vNN/. Each prompt needs iteration id, option id, role, difference axis, prompt, negative prompt, and reject criteria. - Generate or collect assets through the host repo's approved image-generation path.
- Create or update the HTML direction map as the first section of
vNN-<slug>.html. - Put preview cards under the map with anchors matching tree node ids.
- Capture feedback as first-class nodes in the tree and in
process/vNN/...log.md. - Create the next iteration from feedback:
feedback -> interpretation -> new branch -> asset -> decision. - Mark selected, rejected, paused, and next branches explicitly.
Iteration Cluster Contract
Every style-search HTML page starts with a direction graph that answers:
- Where did this visual direction come from?
- What was the brief for each iteration?
- Which option leaf nodes were shown for that iteration?
- Which option was selected?
- What changed in the next brief because of that selection?
- What was carried forward, dropped, or turned into output?
Use skills/art-director/assets/style-search-map-template.html as the starter preview when the host repo has no renderer yet. If a reusable host-repo renderer exists, record its path in the run log and use it as the local reference. A typical reusable setup is:
- proof page:
art-direction/<project>/vNN-dagre-full.html; - shared renderer/data shape:
art-direction/<project>/vNN-style-search-pages.js; - shared visual style:
art-direction/<project>/vNN-style-search-page.css; - graph runtime: a pinned dagre runtime, for example
@dagrejs/dagre@1.1.8, when external network use is allowed by the host repo.
Serve preview pages through a local server:
python3 -m http.server 8080Then open http://127.0.0.1:8080/.... If the preview depends on a CDN and the network is blocked, report the blocker and keep the run in blocked or needs-local-runtime status. Store third-party library source only when the host repo policy explicitly allows vendored dependencies.
Dagre Rendering Contract
Recommended solution: dagre computes the graph layout; the page renders the graph as static HTML/SVG with a detail panel.
Responsibilities:
- dagre: node coordinates and edge routing for the directed decision graph;
- page JavaScript: graph schema normalization, DOM/SVG rendering, click handlers and validation;
- HTML/CSS: readable node cards, status colors, horizontal scroll and detail panel;
- detail panel: shows the selected node's body text and image asset when available.
Use manual SVG layout only as a fallback or baseline. Use Mermaid for docs and quick sketches only when dense style-search history is not required.
Required graph semantics:
- root idea node;
- iteration checkpoint nodes;
- option nodes for each iteration;
- selected option node connects to the next checkpoint;
- unselected options remain leaves;
- production output nodes attach to the producing checkpoint or selected option;
- visible labels use the iteration id, for example
V8-C, so later readers can map preview cards back to the run log.
Cluster schema:
{
"id": "v8",
"title": "V8 - Selected direction expansion",
"brief": "Expand user-liked directions from V7.",
"looked_at": "V8-A, V8-B, V8-C, V8-D",
"selected": "v8-c",
"parent": "v7",
"carry_forward": "V8-C: sharp typography, high-contrast palette, warning label rhythm.",
"drop": "Soft abstract metaphors as the main cover style.",
"feedback": {
"target": "v8-c",
"safe_summary": "Reviewer wants the sharp typography direction pushed further.",
"interpretation": "Intensify type scale and warning-label rhythm in V9."
},
"options": [
{"id": "v8-a", "title": "V8-A - Soft sculpture system", "status": "shown"},
{"id": "v8-c", "title": "V8-C - Brutal manifesto", "status": "selected"}
],
"preview": "assets/v8-selected-directions/c-brutal-typography.png",
"next": "v9",
"terminal_status": "needs-feedback"
}Required cluster fields:
idtitlebrieflooked_atoptionsselectedparentwhen the cluster continues or forks a prior clustercarry_forwarddropfeedbackwhen reviewer feedback existspreviewwhen a representative image existsnextunless the cluster is terminal
Required option statuses:
selectedrejectedshownoutput
Terminal status values:
selected-finalneeds-feedbackblockedneeds-local-runtimereset
HTML Preview Rules
- The decision graph is the first visible section after the header.
- Every iteration cluster shows brief, looked-at variants, options, carry-forward, and dropped motifs.
- Option leaves render as visible nodes attached to their iteration checkpoint.
- Selected option nodes are highlighted and visually connect to the next iteration cluster.
- Production outputs attach to the cluster that produced them.
- The page includes a compact legend for status colors.
- The page includes a current selection block when a direction is selected.
- The page includes a next iteration block when feedback has created follow-up branches.
- The page is static HTML and is previewed through a local server.
- Every node with a representative image opens that image in a detail panel or links to a visible preview card.
Avoid duplicating the same graph nodes immediately below as cards or a second table. Add a legend, selected detail, or next-iteration block only when it answers a different question.
Feedback Loop
When feedback arrives after a preview:
- Record a redacted feedback summary in
process/vNN/...log.md. - Quote raw feedback only when the source is safe for the repo visibility and the user has allowed it.
- Add or update the iteration cluster's
feedback,carry_forward, anddropfields. - Convert the feedback into 1-3 direct design operations: keep, drop, intensify, soften, merge, split, resize, reframe, change format, change reference.
- Create the next iteration cluster with the changed brief and option leaves.
- Generate the next branch set after the new cluster is visible in the HTML map.
Comment log format:
## Feedback - YYYY-MM-DD HH:MM
Target node: `v12-03-operating-cards`
Safe summary:
> ...
Interpretation:
- Keep:
- Drop:
- Next branch:Branch Planning Rules
A good branch set varies real design architecture:
- layout system;
- palette role;
- typography behavior;
- image medium;
- narrative metaphor;
- artifact format;
- density;
- target platform;
- production layer: bitmap image, HTML/CSS overlay, carousel, comic, video storyboard, or another host-approved format.
Weak branch sets vary only color, mood words, or superficial adjectives.
Diversity gate:
- Use 3-8 option leaves.
- Cover at least three materially different axes across the set.
- Fail the branch plan if options differ only by palette, mood words, or adjectives.
Selection Rules
Record selection as an operational decision:
- selected node id;
- why it won;
- what stays invariant;
- what must be removed;
- where the direction will be used next;
- required next asset or generation pass.
If the reviewer rejects the whole pass, make a new root child named reset and record the failure mode before generating again.
Privacy And Source Rules
- Keep private chats, private screenshots, auth material, billing records, customer records, infrastructure details, and non-public repository links out of prompts, previews, READMEs, and public logs.
- Use repo-relative paths in run logs and examples.
- Replace internal project names with generic examples before publishing a reusable skill.
- Public references need source URL, author or owner, access date, and a usage boundary.
- Exact protected characters, brands, logos, or artist-style references need explicit rights, license, or editorial-risk notes in the host repo run log.
Validation
Run the narrow checks that match the pass:
test -f art-direction/<project>/vNN-<slug>.html
rg -n "dagre|style-tree-data|cluster|graph|node-details" art-direction/<project>/vNN-<slug>.html
test -f art-direction/<project>/process/vNN/YYYY-MM-DD-vNN-<slug>-log.md
test -f art-direction/<project>/prompts/vNN/YYYY-MM-DD-vNN-<slug>-prompts.md
rg -n "feedback|carry_forward|drop|selected|terminal_status" art-direction/<project>/vNN-<slug>.html
git diff --checkIf the preview uses an external CDN, also check the pinned runtime:
curl -I -L --max-time 10 https://cdn.jsdelivr.net/npm/@dagrejs/dagre@1.1.8/dist/dagre.min.jsFor local visual review:
python3 -m http.server 8080Then open http://127.0.0.1:8080/... and verify:
- the tree renders;
- every tree node with an asset links to a visible card;
- clicking a node opens its description and image in the detail panel;
- node count and edge count are greater than zero;
- the selected path exists;
- every referenced asset path resolves or is marked as pending;
- selected, rejected, and next states are visible;
- the current selection and next iteration are understandable from files.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

