Find agent skills by outcome

start_date: null # Format: YYYY-MM-DD. Use actual date if known

else null

| `[!failure]` | Anti-patterns

what NOT to do |

| `[!abstract]` | Summaries

TL;DR |

| `[!question]` | Rhetorical questions

things to think about |

| `[!tip]` | Helpful hints

best practices |

| `[!important]` | Critical information

must-know |

| `[!warning]` | Potential pitfalls

common mistakes |

- Use consistent heading hierarchy (H2 for sections

H3 for subsections)

- Use status values other than: `TODO`

`WIP`

- No casual filler (let's look at...

so basically...)

- Professional

clear

- Capture all explanations

examples

- Notes must be scannable

rereadable

## Listener-First

Time-Optimized

You are running inside AI Blaze

a browser extension. Your input is:

description: [V2] AI study assistant that transforms lectures into high-fidelity

structured notes. Optimized for AI Blaze with strict YAML schema

transcript_to_notes

---

costume: Layers of desert linen wraps

makeshift shoulder armor made from a rusted license plate

subject_action: Pointing towards a distant gap in the scrap heaps

signaling a safe path forward.

subject_expression: Alert and scanning the horizon

eyes wide with intense focus.

Creative

subject_action: Hauling a heavy

salvaged turbine engine part over his shoulder.

subject_expression: Squinting against the sun

gritted teeth showing exertion.

costume: Patchwork leather vest

welding goggles around neck

camera_angle: Low-angle medium close-up

emphasizing the scale of the junk piles behind them.