CLI Tools Guide

@astro-minimax/cli provides a comprehensive command-line toolkit for blog project management and AI content processing. This guide covers all available commands.

Installation

In the monorepo / example blog, the CLI is already wired in via a workspace dev dependency. If you’re using it in your own project separately, you can install it manually:

pnpm add -D @astro-minimax/cli

Use via astro-minimax command or pnpm run shortcuts.

Create a New Blog

npx @astro-minimax/cli init my-blog

Generates a complete blog project with config files, sample content, and AI toolchain.

Post Management

Create Posts

pnpm run post:new -- "Post Title"
pnpm run post:new -- "Chinese Title" --lang=zh
pnpm run post:new -- "Tutorial" --category="Tutorial/Frontend"

Automatically creates Markdown files with frontmatter in the appropriate language directory.

List Posts

pnpm run post:list

Shows all posts sorted by date (including subdirectories), distinguishing published posts and drafts.

Post Statistics

pnpm run post:stats

Displays Chinese/English post count statistics.

AI Content Processing

Requires environment variables:

# .env
AI_API_KEY=your-api-key
AI_BASE_URL=https://api.openai.com  # optional
AI_MODEL=gpt-4o-mini                 # optional

Process Articles

pnpm run ai:process                          # Process all articles (summaries + SEO)
pnpm run ai:process -- --force               # Force reprocess
pnpm run ai:process -- --slug=en/my-post     # Process specific article
pnpm run ai:process -- --lang=en             # English articles only
pnpm run ai:process -- --recent=5            # Process recent 5 articles
pnpm run ai:process -- --dry-run             # Preview mode

Generate Summaries

pnpm run ai:summary

Generate SEO Metadata

pnpm run ai:seo

AI Quality Evaluation

pnpm run ai:eval                                    # Evaluate local server
pnpm run ai:eval -- --url=https://your-blog.com     # Evaluate production
pnpm run ai:eval -- --category=no_answer             # Evaluate specific category
pnpm run ai:eval -- --verbose                        # Detailed output

Evaluation is based on datas/eval/gold-set.json golden test set, automatically checking:

• Non-empty response
• Topic coverage
• Forbidden claims not present
• Markdown link existence
• Answer pattern matching

Evaluation report is saved to datas/eval/report.json.

Author Profile

Complete Build

pnpm run ai:profile:build

This is the retained canonical entrypoint for the author profile pipeline. Its output is then used by the runtime knowledge bundle and AI chat flow.

Canonical Build Entry

pnpm run ai:profile:build   # Run the retained author profile pipeline

If you also need the fact registry or extension system, continue with:

pnpm run ai:facts:build
pnpm run ai:facts:validate
pnpm run ai:extensions:status

AI Fact Registry

AI extracts verified facts from blog content and injects them into prompts to reduce hallucinations:

# Build fact registry
pnpm run ai:facts:build

# Validate facts
pnpm run ai:facts:validate

# Check status
pnpm run ai:facts:status

AI Extension System

The extension system provides custom context sections, semantic fallback rules, and other AI enhancements:

# Build extensions
pnpm run ai:extensions:build

# Validate extensions
pnpm run ai:extensions:validate

# Check extension status
pnpm run ai:extensions:status

# Load extensions
pnpm run ai:extensions:load

Git Hooks

Install Hooks

pnpm run hooks:install

Installs Husky Git hooks with auto pubDatetime/modDatetime fill on pre-commit.

Uninstall Hooks

pnpm run hooks:uninstall

Check Hook Status

pnpm run hooks:status

Data Management

View Status

pnpm run data:status

Shows status of all data files, processing counts, and last update time.

Clear Cache

pnpm run data:clear

Clears AI-generated summaries, SEO data, author profiles, and other cache files. Does not delete evaluation reports.

Command Reference