@dedesfr/prompter 0.8.8 → 0.8.9

package/src/commands/init.ts

@@ -70,6 +70,67 @@ export class InitCommand {
         return choices;
     }
 
+    private getCategorizedSkillChoices(availableSkills: SkillMetadata[], currentSkillNames: string[]): any[] {
+        const categories = [
+            {
+                name: '📋 Planning & Strategy',
+                skills: ['project-orchestrator', 'feature-planner', 'prompter-workflow', 'prompter-specs']
+            },
+            {
+                name: '🎨 Design & UI/UX',
+                skills: ['ui-ux-pro', 'design-system-generator', 'design-md']
+            },
+            {
+                name: '⚙️ Development & Code Review',
+                skills: ['code-review', 'laravel-code-review', 'mcp-builder']
+            },
+            {
+                name: '📝 Documentation',
+                skills: ['doc-builder', 'document-translator', 'agents-md-generator', 'meeting-notes']
+            },
+            {
+                name: '✨ Content & Productivity',
+                skills: ['gamma-builder', 'enhance-prompt']
+            }
+        ];
+
+        const categorized = new Set<string>();
+        const choices: any[] = [];
+
+        for (const category of categories) {
+            const skillsInCategory = category.skills
+                .map(name => availableSkills.find(s => s.name === name))
+                .filter((s): s is SkillMetadata => !!s);
+
+            if (skillsInCategory.length === 0) continue;
+
+            choices.push(new Separator(chalk.bold.cyan(category.name)));
+            for (const skill of skillsInCategory) {
+                categorized.add(skill.name);
+                choices.push({
+                    name: `  ${skill.name}`,
+                    value: skill.name,
+                    checked: currentSkillNames.includes(skill.name)
+                });
+            }
+        }
+
+        // Uncategorized skills
+        const uncategorized = availableSkills.filter(s => !categorized.has(s.name));
+        if (uncategorized.length > 0) {
+            choices.push(new Separator(chalk.bold.cyan('🔧 Other')));
+            for (const skill of uncategorized) {
+                choices.push({
+                    name: `  ${skill.name}`,
+                    value: skill.name,
+                    checked: currentSkillNames.includes(skill.name)
+                });
+            }
+        }
+
+        return choices;
+    }
+
     async execute(options: InitOptions = {}): Promise<void> {
         const projectPath = process.cwd();
         const isReInitialization = await PrompterConfig.prompterDirExists(projectPath);
@@ -177,11 +238,8 @@ export class InitCommand {
 
                 const selectedSkillNames = await checkbox({
                     message: 'Select skills to install:',
-                    choices: availableSkills.map(skill => ({
-                        name: `  ${skill.name}`,
-                        value: skill.name,
-                        checked: currentSkillNames.includes(skill.name)
-                    }))
+                    choices: this.getCategorizedSkillChoices(availableSkills, currentSkillNames),
+                    pageSize: 20
                 });
                 selectedSkills = availableSkills.filter(s => selectedSkillNames.includes(s.name));
             } catch (error) {

package/docs/product-spec.md

@@ -1,151 +0,0 @@
-Product Specification Document
-
-### 1. Executive Summary
-- Product name and tagline: **Prompter CLI** — “Enhance prompts directly in your AI coding workflow.”
-- Core value proposition: Turn rough ideas into **structured, professional prompt specifications** and install tool-specific slash-command workflows in local repos.
-- Primary target users: Developers/teams using AI coding assistants who want standardized prompt enhancement workflows.
-- Key differentiators: **Multi-tool workflow generation**, **local-first** (filesystem-only), and **managed template updates** using embedded markers.
-
-### 2. Product Overview
-- **Problem Statement:** AI prompt quality is inconsistent; teams need a repeatable way to transform rough ideas into high-quality specs and make them available as slash commands in their tools.
-- **Solution:** A CLI that initializes a project with Prompter context, generates slash-command workflow files for supported AI tools, and helps list enhanced prompts.
-- **Success Metrics:**
-  - Successful initialization rate (`prompter init` completes).
-  - Number of workflow files created/updated.
-  - Number of enhanced prompts listed (`prompter list` output count).
-- **Scope:**
-  - **Included:** CLI commands (`init`, `update`, `list`), file templates for workflows, project context file, and detection of enhanced prompts.
-  - **Excluded:** Actual prompt enhancement execution (done by the AI tool), remote services, authentication, telemetry.
-
-### 3. User Personas
-- **Primary: AI-Enabled Developer**
-  - Goals: Quickly bootstrap AI prompt workflows; keep workflows up to date.
-  - Pain points: Inconsistent prompt specs; manual setup across tools.
-- **Secondary: Team Lead/Architect**
-  - Goals: Standardize AI prompt practices across team repos.
-  - Pain points: Tool fragmentation and drift in prompt instructions.
-- **Use Case Scenarios:**
-  - New repo bootstrap → run `prompter init` to scaffold workflows.
-  - Existing repo update → run `prompter update` to refresh managed sections.
-  - Review prompt history → run `prompter list` to see enhanced specs.
-
-### 4. Feature Specifications
-
-**Feature Name:** CLI Initialization
-- **Description:** Creates `prompter/` directory, `prompter/project.md`, optional `AGENTS.md`, and tool-specific workflow files.
-- **User Flow:**
-  1. User runs `prompter init`.
-  2. CLI checks if `prompter/` exists; if so, exits with warning.
-  3. User selects tools (interactive checkbox) or passes `--tools`.
-  4. CLI writes templates and workflow files.
-  5. CLI prints next steps.
-- **Acceptance Criteria:**
-  - `prompter/` directory created with `project.md`.
-  - `AGENTS.md` created only if missing.
-  - Tool workflow files created under tool-specific paths.
-  - Success message printed.
-- **Priority:** Critical
-- **Dependencies:** `src/commands/init.ts`, `src/core/config.ts`, templates in `src/core/templates/*`, configurators in `src/core/configurators/slash/*`.
-
-**Feature Name:** Workflow Update
-- **Description:** Updates managed template sections in existing workflow files for all supported tools.
-- **User Flow:**
-  1. User runs `prompter update`.
-  2. CLI checks for `prompter/`; if missing, exits with error.
-  3. For each tool configurator, updates files containing managed markers.
-  4. Outputs updated count.
-- **Acceptance Criteria:**
-  - If `prompter/` absent, prints error and sets exit code.
-  - If managed markers missing, logs failure for that file.
-  - Updates only managed section between markers.
-- **Priority:** High
-- **Dependencies:** `src/commands/update.ts`, `src/core/configurators/slash/base.ts`, `src/core/config.ts`.
-
-**Feature Name:** Enhanced Prompt Listing
-- **Description:** Lists existing enhanced prompts under `prompter/<slug>/enhanced-prompt.md`.
-- **User Flow:**
-  1. User runs `prompter list` (or `--json`).
-  2. CLI scans `prompter/` for directories containing `enhanced-prompt.md`.
-  3. Outputs list and optional JSON structure.
-- **Acceptance Criteria:**
-  - JSON output when `--json` used.
-  - Empty state messaging when no prompts found.
-  - Sorted list by slug.
-- **Priority:** Medium
-- **Dependencies:** `src/commands/list.ts`, `src/core/config.ts`.
-
-**Feature Name:** Tool-Specific Workflow Generation
-- **Description:** Creates or updates slash-command files for supported AI tools.
-- **User Flow:**
-  1. CLI picks tool configurator based on user selection.
-  2. Writes file with frontmatter (if applicable) and managed markers.
-- **Acceptance Criteria:**
-  - Files created under correct tool paths.
-  - Body contains managed markers from `PROMPTER_MARKERS`.
-- **Priority:** High
-- **Dependencies:** `src/core/configurators/slash/*`, `src/core/templates/slash-command-templates.ts`.
-
-### 5. Technical Architecture
-- **System Overview:** Local CLI that reads/writes repository files and generates prompt templates. No server-side components.
-- **Technology Stack:**
-  - **Language:** TypeScript → compiled to Node.js ES Modules (`src/` → `dist/`).
-  - **Runtime:** Node.js >= 20 (`package.json`).
-  - **Libraries:** `commander`, `@inquirer/prompts`, `chalk`.
-- **Data Models:**
-  - `EnhancedPrompt` (in-memory): `{ id, path, createdAt }` in `src/commands/list.ts`.
-  - `ToolChoice` and `SUPPORTED_TOOLS` in `src/core/config.ts`.
-- **APIs & Integrations:** No network APIs. Integrates with tool-specific file conventions.
-- **Infrastructure:** Local filesystem only. CLI entry is `bin/prompter.js` which loads `dist/cli/index.js`.
-
-**Architecture Diagram (Mermaid):**
-```mermaid
-flowchart TD
-  CLI[prompter CLI] -->|init| FS[(Local Filesystem)]
-  CLI -->|update| FS
-  CLI -->|list| FS
-  FS -->|workflow files| Tools[AI Tools: Claude/Codex/etc.]
-  Templates[Templates] --> CLI
-```
-
-### 6. Non-Functional Requirements
-- **Performance:** Fast file IO; scans only `prompter/` directory.
-- **Security:** No authentication; operates on local filesystem; no data transmission.
-- **Scalability:** Handles many prompt folders; linear scan in `prompter/`.
-- **Reliability:** Uses guarded checks for existing directories; safe updates via marker boundaries.
-- **Compliance:** None specified. ⚠️ ASSUMPTION: Local-only operation minimizes compliance needs.
-
-### 7. User Interface Specifications
-- **Navigation Structure:** CLI commands:
-  - `prompter init`
-  - `prompter update`
-  - `prompter list`
-- **Key Screens/Pages:** Terminal output with colored status messages (`chalk`).
-- **Design Patterns:**
-  - Interactive selection via checkbox prompt.
-  - Managed template sections via markers: `<!-- prompter-managed-start -->` and `<!-- prompter-managed-end -->`.
-- **Responsive Behavior:** N/A (CLI).
-
-### 8. Integration Requirements
-- **Third-Party Services:** None.
-- **APIs Consumed:** None.
-- **APIs Provided:** CLI commands only.
-- **Webhooks/Events:** None.
-
-### 9. Data Specifications
-- **Data Models:**
-  - `prompter/project.md` — project context template.
-  - Workflow files per tool, e.g., `.codex/prompts/prompter-enhance.md`.
-- **Data Flow:** User input → CLI templates → filesystem output → AI tool reads workflow file.
-- **Storage Requirements:** Low; markdown text files only.
-- **Data Privacy:** No PII processing; files are local. ⚠️ ASSUMPTION: Enhanced prompts may include sensitive text, handled by repo policies.
-
-### 10. Current Limitations & Technical Debt
-- No validation of `--tools` inputs; invalid tool IDs are ignored silently (no error path in `src/commands/init.ts`).
-- `prompter update` updates all known tools, even if not used in the project.
-- No tests present (no `test/` files found).
-- Update requires markers; files without them error (`src/core/configurators/slash/base.ts`).
-
-### 11. Assumptions & Inferences
-- ⚠️ ASSUMPTION: Prompter is intended to be used as a **global CLI** (`npm install -g`), inferred from README and `package.json` `bin` mapping.
-- ⚠️ ASSUMPTION: Enhanced prompts are generated by the AI assistant following the workflow, not by this CLI.
-- ⚠️ ASSUMPTION: Users will maintain versioned workflow files in their repos and accept managed updates.

package/docs/tasks.md

@@ -1,3 +0,0 @@
-COMMAND_NAME = ['apply', 'archive', 'proposal']
-
-i want to add new command for {COMMAND_NAME}. you can check the prompt in prompt folder here with name match {COMMAND_NAME}.md