AI Agents

All agents follow the agent pattern and use prompts stored in the database.

Agent Overview

Agent	Model	Purpose
Brief Generator	Sonnet	Structures article outlines from backlog items, injects trusted sources
Researcher	Sonnet	Gathers evidence via PubMed, web search, vault notes
Writer	Sonnet	Produces full article drafts
Factuality Checker	Haiku	Validates medical claim accuracy
SEO Optimizer	Haiku	Checks keyword usage, heading structure, snippets
Style Checker	Haiku	Evaluates readability and tone
Synthesis	Haiku	Reconciles quality gate feedback into rewrite instructions
Triage	Haiku	Classifies research news relevance
Digest Writer	Sonnet	Summarizes weekly research findings
Readability	Haiku	Scores readability metrics (Flesch, sentence length)
Image Generator	Haiku	Composes DALL-E prompts from article context, generates featured images
Cornerstone Updater	Sonnet	Integrates new content into cornerstone articles

Model Selection

• Sonnet is used for tasks that require creative writing or complex reasoning (research, drafting, cornerstone updates)
• Haiku is used for classification, scoring, and structured evaluation tasks (quality gates, triage, synthesis)

All prompts are versioned in the database and managed via Alembic data migrations. See Manage Prompts for details.

Language Support

The Factuality Checker supports both English and German content with language-specific prompts (v3) for improved cultural and linguistic accuracy.

Trusted Sources

The Brief Generator queries the trusted_sources table for active sources filtered by article language and injects them as a <trusted_sources> XML block in the user message. This seeds key_sources_to_consult with authoritative sources (Cochrane, NICE, ATA, etc.) without restricting the researcher's search freedom. Sources with language="both" (scientific) are included for all articles; language-specific sources (medical journalism) are filtered by the backlog item's language.

Agent Modules

Each agent is defined in src/agents/ with a corresponding module:

src/agents/
  base.py                  # call_agent() -- shared invocation logic
  schemas.py               # Pydantic models for all agent I/O
  brief_generator.py       # generate_brief()
  researcher.py            # research_topic()
  writer.py                # write_article()
  factuality_checker.py    # check_factuality()
  seo_optimizer.py         # optimize_seo()
  style_checker.py         # check_style()
  synthesis.py             # synthesize_feedback()
  triage.py                # triage_item()
  digest_writer.py         # write_digest()
  cornerstone_updater.py   # update_cornerstone()
  readability.py           # check_readability()
  image_generator.py       # generate_featured_image()
  cost_tracker.py          # Token usage tracking