Quick Reference
Purpose
The research protocol provides comprehensive web research capabilities for aStory blog content generation, including source verification, media collection, and citation management.
Core Features
Multi-Stage Web Search:
- Primary search targeting the core topic directly
- Secondary search expanding into related concepts
- Tertiary search targeting specific technical details
- Verification search to confirm key claims
Media Collection:
- Extract images from verified sources with license tracking
- Collect YouTube videos with timestamp analysis
- Usage plan mapping resources to content sections
Source Verification:
- Verify URL accessibility via WebFetch
- Verify content recency (updates for sources older than 6 months)
- Cross-verify factual accuracy from multiple sources
- Evaluate author credibility
Research Brief Output
The research process produces a structured brief that includes:
- Topic summary with key findings
- Numbered source list with URL, type, and relevance rating
- Image collection with URL, description, and license information
- Video collection with YouTube links, timestamps, and usage plan
- Section mapping showing the resources supporting each content section
- Formatted citations for the references section
Implementation Guide
Research Workflow
Research follows a 5-stage process that builds from broad search to specific resource allocation.
Stage 1 - Topic Analysis:
- Review the topic to identify core concepts
- Identify related technologies and authoritative sources
- Determine search strategy based on content type
Stage 2 - Execute Web Search:
- Perform multiple searches with varied query formulations
- Target different aspects of the topic
- Collect and deduplicate results
Stage 3 - Source Verification:
- Evaluate each potential source for credibility, recency, and relevance
- Only verified sources proceed to the collection stage
- Track source types and prioritize official documentation
Stage 4 - Media Collection:
- Extract images and YouTube videos from verified sources
- Track license information for all media
- Document context and planned usage
Stage 5 - Brief Compilation:
- Organize all resources into a structured brief
- Creating section mapping with resource allocation
- Formatting citations for reference sections
Source priority hierarchy
Official documentation (highest priority):
- Anthropic documentation, official GitHub repositories, company blog posts
- Most authoritative and accurate sources
Established technical publications (second priority):
- Recognized technical blogs, conference papers, peer-reviewed content
- Strong credibility with editorial oversight
Community resources (third priority):
- Highly upvoted Stack Overflow answers
- GitHub issues with official responses
- Tutorial sites with verified accuracy
Personal blog posts (fourth priority):
- Require additional verification before inclusion
- May contain valuable insights but require cross-verification
News articles (fifth priority):
- Provide context but rarely offer technical depth
- Used for timelines and industry context
Verification requirements
All sources must meet verification criteria before inclusion:
URL accessibility:
- WebFetch tool verifies that each URL returns valid content
- Broken links are discarded or replaced
Content recency:
- Sources older than 6 months require checking for updated versions
- Verify technical versions and release dates
Factual accuracy:
- Key claims must appear in at least one additional source
- Cross-reference statistics and benchmarks
Author credibility:
- Give higher weight to known experts and official representatives
- Anonymous or unknown authors require stronger cross-verification
For detailed search protocols, see web-search-protocol.md .
Advanced patterns
Image collection protocol
Image collection follows specific guidelines to ensure proper media usage:
Preferred images:
- Architecture diagrams from official documentation
- Annotated screenshots demonstrating features
- Infographics explaining concepts
- Comparison visualizations
License tracking:
- Prefer Creative Commons licenses
- Unknown licenses require an image URL pointing to the source page rather than the direct image file
- Document the license terms for each image
Context documentation:
- Explain where the image was found
- Describe what it depicts
- Clarify how it supports the content
- Enables creation of appropriate alt text and captions
Usage planning:
- Assign each image to a specific content section
- Prevent multiple images from competing for the same position
- Balance image distribution across the content
For detailed media collection procedures, see media-collection.md .
YouTube Collection Protocol
YouTube videos provide valuable supplementary content when properly curated:
Preferred videos:
- Official product demos
- Conference presentations by recognized experts
- Tutorial walkthroughs of relevant features
- Technical deep dives from established channels
Timestamp analysis:
- Identify specific segments relevant to the topic
- Record timestamp ranges rather than entire videos
- Allow readers to jump to the relevant section
Embedding plan:
- Use the YouTubeCard component for video embeds
- Assign each video to a specific content location
- Provide context explaining why the video supports the section
Research Brief Structure
The research brief follows a standardized format that downstream agents can parse:
Collection Summary:
- Topic identifier
- Number of sources, images, and videos collected
- Search strategy overview
Sources Section:
- Each verified source with ID, URL, and title
- Type classification (official, publication, community, etc.)
- Relevance rating (high, medium, low)
- Key points extracted from the content
Images Section:
- Each collected image with URL and alt text
- Context description
- Planned usage location
- License information
Videos Section:
- Each YouTube video with URL, title, and length
- Timestamp notes for relevant segments
- Planned usage and rationale
Section Mapping:
- Assign resources to content structure
- Show which sources support the introduction
- Indicate where media embeds should appear
Citations Section:
- Pre-formatted entries for the references section
- Follow consistent citation standards
Module Reference
Search Protocol:
- web-search-protocol.md - Multi-step search strategy, query construction, result filtering
Media Collection:
- media-collection.md - Image extraction, YouTube curation, license tracking, usage planning
Works Well With
Skills:
- writing-standards - Apply quality rules to content generated from the research brief
- personas - Consume the research brief to generate content in a characteristic style
- anti-ai-validator - Verify that source citations meet quality standards
Commands:
- post - Orchestrate the full workflow from research to writing to publishing
Quality Standards
Source Verification Requirements
All sources must meet the following criteria:
- URL is valid and returns accessible content via WebFetch
- Content recency appropriate for the technical version being discussed
- Facts cross-verified from at least one additional independent source
- Author credibility assessed and documented
Media verification requirements
Images:
- All image URLs valid and accessible
- License information fully documented
- Context description provides full details
- Usage plan assigned to specific sections
Videos:
- All YouTube URLs valid and accessible
- Timestamps accurate and relevant
- Embed locations appropriately planned
Brief completeness requirements
Section mapping:
- All major content sections have assigned resources
- No orphan resources without section assignment
- Citations properly formatted
Brief structure:
- YAML syntax valid when using structured format
- All required sections present
- Counts match actually collected items
Version: 2.0.0 Last Updated: 2026-01-06 Category: Content Preparation Type: Research and Verification Status: Production Ready