> ## Documentation Index
> Fetch the complete documentation index at: https://docs.engramlabs.xyz/llms.txt
> Use this file to discover all available pages before exploring further.

# XP & Contribution

> How lessons are scored, weighted, and rewarded

Quality dominates volume by design.

## Base XP

Every closed trade that meets minimum requirements starts at 10 base XP. Multipliers stack multiplicatively from there – up or down.

## Thesis Multipliers

| Condition                   | Multiplier     |
| --------------------------- | -------------- |
| Opening thesis provided     | 2.0x           |
| Closing reflection provided | 1.5x           |
| Both provided               | 3.0x (stacked) |
| Neither provided            | 1.0x           |

## Outcome Multiplier

| Condition                                | Multiplier |
| ---------------------------------------- | ---------- |
| Profitable + thesis confirmed            | 1.3x       |
| Profitable + thesis didn't play out      | 1.0x       |
| Unprofitable + sound thesis + reflection | 1.0x       |
| Unprofitable + no reflection             | 1.0x       |

A trade that loses money but demonstrates sound reasoning earns the same outcome multiplier as a lucky win. No penalty for disciplined losses.

## Quality Multipliers (LLM-scored nightly)

| Condition                                            | Multiplier  |
| ---------------------------------------------------- | ----------- |
| Thesis specificity (references verified data points) | up to 1.5x  |
| Invalidation criteria stated                         | 1.2x        |
| Reflection references the actual entry-exit diff     | 1.3x        |
| Cross-handler uniqueness                             | 1.0x - 0.5x |

The LLM cross-references thesis text against the actual 140-field snapshot values. Recycled theses – copy-pasted from previous trades or generated as generic AI filler – are detected and scaled down.

## Context Richness

| Condition                                                                                            | Multiplier |
| ---------------------------------------------------------------------------------------------------- | ---------- |
| Meaningful context diff (price moved >2%, funding flipped, cohort bias shifted >5pp, OI changed >5%) | 1.2x       |
| Thin context diff (market flat, trade too short)                                                     | 0.8x       |

## In Practice

A high-quality lesson – thesis referencing specific data points, invalidation criteria, profitable with thesis confirmed, meaningful diff, reflection addressing what changed – earns \~110 XP.

A low-effort trade – no thesis, no reflection, thin diff – earns \~8 XP.

That's a 13x quality gap. By design.

## Anti-Gaming

| Rule                                | Result                    |
| ----------------------------------- | ------------------------- |
| Hold time below 90 minutes          | 0 XP, no lesson generated |
| Trade size below \$1,000 notional   | 0 XP, no lesson generated |
| More than 10 scoring trades per day | 0.5x on additional trades |

Continuous pattern detection catches wash trading, rapid open/close cycles, and coordinated manipulation.

## Visibility & Timing

Structural multipliers (thesis provided, reflection provided, outcome, context diff) are visible immediately at trade close. The handler knows what they earned and why.

LLM quality scores land the next morning via nightly batch. This prevents real-time gaming against the evaluator.

## Competitions

Structured around XP, not P\&L. Season-based leagues with cumulative XP rankings, weekly sprints, and team competitions.

## Ownership

Cumulative quality-weighted XP tracks each handler's contribution to the intelligence layer. The most thoughtful contributors – those who consistently provide reasoning alongside trades – build the largest track record. How XP translates to long-term ownership, governance, or economic rights will be defined as the network matures. The ledger starts recording from day one. Intelligence production, not capital, determines standing.