All LLM prompt templates used by the AI summary feature. These files are loaded at runtime by src/lib/llm-service.ts and must be deployed with the application.
| File | Purpose | Used by | Modes |
|---|---|---|---|
bullets.md |
Bullet-point summary (10-15 bullets with YouTube timestamp links) | All 3 LLMs | Bullets |
narrative.md |
Flowing essay (Opening, Key Ideas, Practical Takeaways, Closing) | All 3 LLMs | Narrative |
technical.md |
Structured extraction (Tools, Workflows, Tips, Metrics) | All 3 LLMs | Technical |
fallback.md |
Minimal fallback prompt used when a template file fails to load | All 3 LLMs | All modes |
Anthropic's Messages API supports a dedicated system parameter that the model treats as high-priority behavioral constraints. The code splits each template into two parts:
- System prompt (
systemparam): Role, Critical Rules, Context, Constraints, Quality Checklist, Final Reminder sections - User message (
messages[0]): Task, Output Structure/Format, Examples, Episode Time Range, Transcript
This separation follows Anthropic's recommended API pattern for separating instructions from content and is handled by buildAnthropicPromptParts() in src/lib/llm-api-helpers.ts. Temperature is set to 0.7.
Gemini receives the entire template + transcript as a single content block via buildFullPrompt(). No system/user split. Temperature is 0.7.
Perplexity receives the entire template + transcript as a single user message via buildFullPrompt(). No system/user split. Temperature is 0.7.
| Mode | Limit | Enforced via |
|---|---|---|
| Bullets | 10-15 bullet points | Critical Rule 5 in bullets.md |
| Narrative | 750-1000 words | Critical Rule 6 in narrative.md |
| Technical | 2000 words max | Critical Rule 7 in technical.md |
All 3 style templates share the same ## Section layout:
| Section | Goes into (Anthropic) | Purpose |
|---|---|---|
## Role |
System prompt | Who the model is |
## Critical Rules |
System prompt | Hard constraints (accuracy, limits) |
## Context |
System prompt | Grounding instructions |
## Task |
User message | What to produce |
## Output Format/Structure |
User message | Formatting rules and examples |
## Constraints |
System prompt | Length and style limits |
## Quality Checklist |
System prompt | Self-verification checklist |
## Final Reminder |
System prompt | Last-priority reinforcement |
For Gemini and Perplexity, all sections are concatenated into a single message.
- Changes to template files apply to all 3 LLMs immediately (no code deploy needed, just file update)
- Keep Critical Rules numbered sequentially within each template
- Test changes against at least 2 different transcripts before considering them final