菜单
Discover and evaluate AI-related Substack newsletters for content curation. Use when the user says "discover sources", "find substacks", "run source discovery", or "find newsletters".
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Push the LLM to reconsider, refine, and improve its recent output.
Facilitate interactive brainstorming sessions using diverse creative techniques and ideation methods. Use when the user says help me brainstorm or help me ideate.
Clinical copy-editor that reviews text for communication issues. Use when user says review for prose or improve the prose
Structural editor that proposes cuts, reorganization, and simplification while preserving comprehension. Use when user requests structural review or editorial review of structure
Analyzes current state and user query to answer BMad questions or recommend the next workflow or agent. Use when user says what should I do next, what do I do now, or asks a question about BMad
Generates or updates an index.md to reference all docs in the folder. Use if user requests to create or update an index of all files in a specific folder
| name | source_discovery |
| description | Discover and evaluate AI-related Substack newsletters for content curation. Use when the user says "discover sources", "find substacks", "run source discovery", or "find newsletters". |
You are a Source Discovery specialist. Your job is to search Substack for AI newsletters matching user-specified keywords or topics, score them, and stage candidates to Notion for review.
Before running any discovery:
python_services/packages/source_discovery/src/substack_discoverer.py to understand current capabilities.python_services/packages/source_discovery/config/discovery_profile.yaml for available topics and keywords..discovery_checkpoint.json exists in the package root, consider using --resume to avoid re-fetching.Run discovery from python_services/packages/source_discovery:
# All topics from config
python -m src.substack_discoverer --all-topics --stage
# Resume interrupted discovery
python -m src.substack_discoverer --all-topics --stage --resume
# Single topic
python -m src.substack_discoverer --topic "Model Updates" --stage
# Custom keywords for a topic
python -m src.substack_discoverer --topic "Case Studies" \
--keywords "AI case study" "enterprise AI" \
--max-results 10 --stage
# Check checkpoint status
python -m src.substack_discoverer --checkpoint-status
# Start fresh
python -m src.substack_discoverer --clear-checkpoint --all-topics --stage
| Flag | Purpose |
|---|---|
--all-topics | Run all topics from config |
--topic "NAME" | Run a single topic |
--keywords ... | Override config keywords |
--max-results N | Limit results per topic |
--stage | Stage passing candidates to Notion |
--resume | Skip already-checked URLs/topics |
--checkpoint-status | Show checkpoint state |
--clear-checkpoint | Reset and start fresh |
| Factor | Points | Criteria |
|---|---|---|
| Topic Relevance | 0-2 | Keywords in name/description/posts |
| Posting Consistency | 0-1 | Post count threshold |
| Subscriber Count | 0-1 | 1K-100K = 1.0 |
| Content Quality | 0-0.5 | >= 3 recent posts |
Min score to stage: 3.0. Priority: High (>=4.0), Medium (>=3.5), Low (<3.5).