implement-ag-playbook
Skill833 starsChanged 9 months agoApache-2.0View on GitHub
Implement a playbook from its PROMPT specification files. Supports both single-framework (PROMPT-REACT.md, etc.) and 3-framework (PROMPT-3-FRAMEWORKS.md) modes.
- Installs packages
In this repository5 other files it gives its agents.
- check-accessibilitySkill
- explain-codeSkill
- fix-github-issueSkill
- implement-github-featureSkill
- update-component-docsSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
16 rules. About 0 are common, 10 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 48
If MODE specified, verify that PROMPT file exists
- Distinctive: line 49
If no MODE specified, ask user which to implement
- Distinctive: line 52
Read the target PROMPT file(s)
- Distinctive: line 66
Create appropriate project(s):
- Very specific: line 99
Check design/ folder for assets beyond mockups
- Very specific: line 101
If assets exist (icons, logos, background images), copy to each project's public/ folder
- Distinctive: line 102
Follow asset instructions in the PROMPT file
- Very specific: line 105
Add custom element recognition for ag-* tags:
- Distinctive: line 117
Import in each project's entry point:
- Very specific: line 124
Add Google Fonts link to each index.html per PROMPT specification
- Very specific: line 127
If CLI-generated code causes errors, relax tsconfig.app.json:
- Distinctive: line 139
Follow the PROMPT specification exactly
- Very specific: line 145
Run npm run build in each project
- Distinctive: line 160
All implementations should behave identically
- Distinctive: line 170
Commit with descriptive message
- Distinctive: line 207
If a playbook is missing PROMPT files, notify the user:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Usage: /implement-ag-playbook PLAYBOOK_NAME [MODE]
Examples:
/implement-ag-playbook onboarding- Implement all available prompts/implement-ag-playbook login 3-frameworks- Implement only PROMPT-3-FRAMEWORKS.md/implement-ag-playbook login react- Implement only PROMPT-REACT.md
Implement the specified playbook based on its existing PROMPT specification files.
Playbook Structure Reference
Complete playbooks have this structure (see v2/playbooks/README.md):
v2/playbooks/[name]/
├── PROMPT-3-FRAMEWORKS.md # Builds → react-example/, vue-example/, lit-example/
├── PROMPT-REACT.md # Builds → react/
├── PROMPT-VUE.md # Builds → vue/
├── PROMPT-LIT.md # Builds → lit/
└── design/ # Shared assets (mockups, icons, images)Single-Framework vs 3-Frameworks differences:
- Different fonts (e.g., Playfair Display vs Merriweather)
- Different social icons (e.g., Apple vs Facebook)
- Different asset filenames (e.g.,
logo.svgvslogo-3-frameworks.svg)
Phase 0: Validation
- Verify playbook exists
- Check
v2/playbooks/$ARGUMENTS/exists - List available PROMPT files:
PROMPT-3-FRAMEWORKS.mdPROMPT-REACT.mdPROMPT-VUE.mdPROMPT-LIT.md
- If MODE specified, verify that PROMPT file exists
- If no MODE specified, ask user which to implement
- Check
- Read the prompt specification
- Read the target PROMPT file(s)
- Extract:
- Required components (for
agnosticui-cli add) - Design specifications
- Font requirements
- Asset copying instructions
- Framework-specific configurations
- Required components (for
Phase 1: Project Scaffolding
Create Vite project(s)
- Navigate to
v2/playbooks/$ARGUMENTS/ - Create appropriate project(s):
For 3-frameworks mode:
npm create vite@latest react-example -- --template react-ts npm create vite@latest vue-example -- --template vue-ts npm create vite@latest lit-example -- --template lit-tsFor single-framework mode:
npm create vite@latest react -- --template react-ts # PROMPT-REACT.md npm create vite@latest vue -- --template vue-ts # PROMPT-VUE.md npm create vite@latest lit -- --template lit-ts # PROMPT-LIT.md- Navigate to
- Install dependencies
- React:
npm install && npm install lucide-react lit - Vue:
npm install && npm install lucide-vue-next lit - Lit:
npm install && npm install lit
- React:
- Initialize AgnosticUI CLI
- For each project:
npx agnosticui-cli init --framework [react|vue|lit] --skip-prompts npx agnosticui-cli add [COMPONENTS from prompt]
Phase 2: Configuration
- Copy design assets (if applicable)
- Check
design/folder for assets beyond mockups - Not all playbooks have assets to copy - some only have mockups (e.g.,
onboarding) - If assets exist (icons, logos, background images), copy to each project's
public/folder - Follow asset instructions in the PROMPT file
- Check
- Vue vite.config.ts
- Add custom element recognition for
ag-*tags:
vue({ template: { compilerOptions: { isCustomElement: (tag) => tag.startsWith('ag-'), }, }, }) - Add custom element recognition for
- Token imports
- Import in each project's entry point:
import './components/ag/styles/ag-tokens.css' import './components/ag/styles/ag-tokens-dark.css' - Font setup
- Add Google Fonts link to each
index.htmlper PROMPT specification
- Add Google Fonts link to each
- TypeScript configuration (Vue)
- If CLI-generated code causes errors, relax
tsconfig.app.json:
"strict": false, "noUnusedLocals": false, "noUnusedParameters": false - If CLI-generated code causes errors, relax
Phase 3: Implementation
- Implement each framework
- Follow the PROMPT specification exactly
- React:
src/App.tsxusing React wrappers + lucide-react - Vue:
src/App.vueusing Vue wrappers + lucide-vue-next - Lit: Custom element using core components + inline SVG icons
- Verify builds
- Run
npm run buildin each project - Fix any TypeScript or build errors
- Test with
npm run dev
- Run
Phase 4: Verification
- Visual verification
- Run each example
- Compare against design mockups in
design/ - Check responsive breakpoints
- Verify component interactions
- Cross-framework consistency
- All implementations should behave identically
- State management works correctly
- Styling matches across frameworks
Phase 5: Commit
- Stage and commit
- Show
git diff --stat - Commit with descriptive message
- WAIT FOR USER APPROVAL
- Show
Key Patterns
Icon Imports
- React:
import { Icon } from "lucide-react" - Vue:
import { Icon } from "lucide-vue-next" - Lit: Inline SVG templates (lucide doesn't work directly)
Component Imports
- React:
./components/ag/[Component]/react/ - Vue:
./components/ag/[Component]/vue/ - Lit:
./components/ag/[Component]/core/
Event Handling
- Selection components emit
selection-changeevents - Event detail:
{ value, checked, selectedValues }
Common Issues & Fixes
| Issue | Fix |
Vue ag-* warnings | Add isCustomElement to vite.config.ts |
| TypeScript strict errors | Relax tsconfig.app.json settings |
| Missing dark mode | Import both ag-tokens.css AND ag-tokens-dark.css |
Button isDisabled prop | Use disabled not isDisabled |
| Timeline connector misalignment | Ensure core component has ::slotted(*) { box-sizing: border-box } |
Incomplete Playbook Warning
If a playbook is missing PROMPT files, notify the user:
Warning: The
$ARGUMENTSplaybook is incomplete.Present: [list files] Missing: [list files]
Complete playbooks should have:
PROMPT-3-FRAMEWORKS.md→react-example/,vue-example/,lit-example/PROMPT-REACT.md→react/PROMPT-VUE.md→vue/PROMPT-LIT.md→lit/Would you like to:
- Implement only the available prompts
- Generate the missing PROMPT files first
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.

