design
Skill0 starsChanged 6 days agoMITView on GitHub
Turns an approved spec into contracts, risks and a gated task plan. Use when the user plans how to build scoped work. Not for investigation (specify).
198 other projects use a version of this.
Adopted by
Most recently taken up by
- irank786/Claude-skillstheir own version, first seen Nov 16, 2026
- GoBeromsu/craft-skillstheir own version, first seen Nov 16, 2026
- david33137/claude-skillstheir own version, first seen Nov 16, 2026
- StitchSlop/claude-plugintheir own version, first seen Nov 16, 2026
- FAHTHIZ/oasistheir own version, first seen Nov 16, 2026
- fatihkan/baditheir own version, first seen Nov 16, 2026
Used with
Projects with it also use banner-design, ui-styling and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
lithStudySkillbanner-design
Design banners for social media, ads, website heroes, creative assets, and print. Multiple art direction options with AI-generated visuals. Actions: design, create, generate banner. Platforms: Facebook, Twitter/X, LinkedIn, YouTube, Instagram, Google Display, website hero, print. Styles: minimalist, gradient, bold typography, photo-based, illustrated, geometric, retro, glassmorphism, 3D, neon, duotone, editorial, collage. Uses ui-ux-pro-max, frontend-design, ai-artist, ai-multimodal skills.
Also in 39 projects that use this
MicrosoftSkillui-styling
Guide for the visual style, structure, and shared building blocks used by Semantic Link Labs interactive UI tools (HTML widgets and anywidget-based widgets). Use this when adding a new interactive UI, modifying an existing one, or adding shared visual components.
Also in 41 projects that use this
lithStudySkillbrand
Brand voice, visual identity, messaging frameworks, asset management, brand consistency. Activate for branded content, tone of voice, marketing assets, brand compliance, style guides.
Also in 44 projects that use this
OpenAnalystIncSkillads-management
Activate for paid advertising campaigns on Google Ads, Meta Ads, LinkedIn Ads, TikTok Ads. Includes ad copywriting, audience targeting, budget optimization, A/B testing, and ROAS tracking. Used by ads-specialist and campaign-manager agents.
Also in 3 projects that use this
OpenAnalystIncSkillai-artist
Write and optimize prompts for AI-generated outcomes across text and image models. Use when crafting prompts for LLMs (Claude, GPT, Gemini), image generators (Midjourney, DALL-E, Stable Diffusion, Imagen, Flux), or video generators (Veo, Runway). Covers prompt structure, style keywords, negative prompts, chain-of-thought, few-shot examples, iterative refinement, and domain-specific patterns for marketing, code, and creative writing.
Also in 3 projects that use this
mayankjain1815-commitsSkillinfra-audit
Reads docker-compose, env files, ORM configs, and connection strings to map current infrastructure. Flags missing layers (cache, queue, analytics) based on observed access patterns. Outputs a structured infrastructure manifest.
Also in 3 projects that use this
- +5 more from OpenAnalystInc
Other versions198 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 198
- nextlevelbuilder/ui-ux-pro-max-skill★ 150ksame name
- Yeachan-Heo/oh-my-codex★ 34ksame name
- tokio-rs/toasty★ 3.3ksame name
- PaulWoitaschek/Voice★ 3.3ksame name
- joye61/pic-smaller★ 2.9ksame name
- oxsecurity/megalinter★ 2.7ksame name
- Infisical/agent-vault★ 2.5ksame name
- static-web-server/static-web-server★ 2.4ksame name
- SethGammon/Citadel★ 927same name
- Ohh-889/skyroc-admin★ 795same name
188 more not listed
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 repository22 other files it gives its agents.
- AGENTS.mdAGENTS.md
- claude-arsenal/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- coverage-gapsSkill
- dep-upgradeSkill
- executionSkill
- explore-ideaSkill
- gate-checkSkill
- githubSkill
- initSkill
- litcriticSkill
- mutmut-reportSkill
- pin-checkSkill
- pypi-releaseSkill
- python-bootstrapSkill
- queue-addSkill
- queue-nextSkill
- queue-statusSkill
- reviewSkill
- session-endSkill
- shipSkill
- specifySkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 175 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 43
Use concrete examples (JSON payloads, SQL migrations, config snippets).
- Very specific: line 57
Keep each task small enough for one commit.
- Distinctive: line 78
After writing the files, draw the dependency graph from the task table and confirm the plan's structure:
- Very specific: line 115
When arsenal/project/<WORKSPACE>/ exists, write the plan to arsenal/project/<WORKSPACE>/plan.md (and the contracts/risks tail to the …
“Common” is measured: kept by at least 6 in 10 of the 175 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
Design Workflow
CANARY: design-loaded-2026-05-19-e08675ccb0a5c932
Owns sections 5–6 of status/specification.md (contracts, risks) and creates status/plan.md (task split). Reads sections 1–4 of status/specification.md to understand what the spec already covers.
Before Step 1 — the spec must be approved. The specify skill's validator
answers it: validate_spec.py --input status/specification.md --require-approved
(in that skill's scripts/).
Anything but exit 0 means stop and hand the spec back for approval, because a plan written before approval is built on a draft.
Steps
Step 1: Define the technical solution
Translate the chosen option into a concrete technical design.
- Architecture overview: how the change fits into the existing system
- Data flow: how data moves through affected services (request/response, events, jobs)
- State changes: what data is created, updated, or deleted and where
- Technology choices: any new libraries, tools, or patterns (justify each)
- What is NOT changing: explicit boundaries to prevent scope creep
Step 2: Define contracts
For every interaction between components:
- API contracts: request/response schemas, status codes, error formats
- Event contracts: message schemas, routing keys, retry policies
- Database changes: schema modifications, migration approach (always forward-compatible)
- Configuration: new env vars, feature flags, deployment parameters
Use concrete examples (JSON payloads, SQL migrations, config snippets).
Step 3: Split into tasks
Break the implementation into ordered, independently testable tasks.
For each task:
- What: specific deliverable
- Where: which files/services
- Dependencies: what must be done first
- Gate:
<metric> <op> <threshold>derived from the spec's success criteria (p95_latency_ms <= 200,line_coverage >= 0.90), in thegate-checkgrammar so it can be checked mechanically; non-numeric only when no number fits. - Tests: file path(s) and
test_<what>_<condition>_<expected_result>names, each with a one-sentence assertion; they are copied into the task payload so the worker writes them failing first. - Estimated effort: Small (< 1h) / Medium (1-4h) / Large (4h+)
Keep each task small enough for one commit. execution records each gate's evidence in the plan's Evidence log, which review and ship audit.
Step 4: Anticipate risks
For each risk:
- What could go wrong: specific failure scenario
- Likelihood: Low / Medium / High
- Impact: Low / Medium / High
- Mitigation: what to do to prevent or handle it
- Rollback plan: how to undo if it goes wrong
Pay special attention to:
- Backwards compatibility (API consumers, data formats)
- Data integrity during migration
- Performance under production load
- Deployment ordering (if multi-service)
Step 5: Produce design output
Load references/template-specification-tail.md when appending contracts and risks (sections 5–6) to status/specification.md. Load references/template-plan.md when creating status/plan.md.
After writing the files, draw the dependency graph from the task table and confirm the plan's structure:
python3 "${CLAUDE_SKILL_DIR}/scripts/sync_plan_graph.py" --input status/plan.md --apply
python3 "${CLAUDE_SKILL_DIR}/scripts/validate_plan.py" --input status/plan.mdThe graph is generated, never hand-drawn; rerun sync_plan_graph.py --apply after any change to Depends. The validator checks shape — the required sections and task-table columns, Gate included — and that every Depends edge is in the graph (a path counts: the graph leaves out edges a longer path implies, and past drawspec's one-flow budget it is split into a phase overview plus one graph per phase). A graph that is not a drawspec fence (an older plan's ASCII art) only warns, with the --apply fix. gate-check's run_gate.py --strict audits the gate values.
Draw any picture as a fenced drawspec block: claude-arsenal:core:init § references/diagrams.md.
Step 6: Publish the annotatable plan
Once the validator passes, generate the reader and give the user the HTML in the same
reply; the plan is what they sign off on, section by section. Name the plan
explicitly, since auto-discovery finds only specs, and send --output-dir with it
(arsenal/project/<WORKSPACE>/ for a workspace plan):
create_reader.py --input status/plan.md --output-dir status # in claude-arsenal/scripts/; uv run --with markdownIt writes <repo>-plan-v<N>-<date>-reader.html and …-annotated.md and prints both paths. The plan
keeps the spec's review record (**Revision**, **Status**, **Revision log**);
seeding tasks or starting execution waits for
validate_plan.py --input <plan> --require-approved. Load
claude-arsenal:core:init § references/annotatable-reader.md when a returned notes
export arrives or when recording approval.
Abbreviation
Abbreviated design = Step 1 (one paragraph) + Step 3 (task list only), where the host repo's CLAUDE.md allows it.
Workspace-aware paths
When arsenal/project/<WORKSPACE>/ exists, write the plan to arsenal/project/<WORKSPACE>/plan.md (and the contracts/risks tail to the workspace's spec.md) instead of status/plan.md. Otherwise use status/ as above. The validator takes the path via --input; point it at whichever file was written.
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.

