Research Stellar study streak rewards

docs/stellar-study-streak-rewards-research.md

@@ -0,0 +1,146 @@
+# Stellar Study Streak Rewards Research
+
+Study Copilot may eventually use Stellar assets to encourage consistent study habits. Rewards can be motivating, but they also add abuse, compliance, and product-risk concerns. This document outlines reward scenarios, Stellar asset options, abuse risks, and an MVP recommendation.
+
+## Research Sources
+
+- Stellar assets: https://developers.stellar.org/docs/learn/fundamentals/stellar-data-structures/assets
+- Stellar accounts: https://developers.stellar.org/docs/learn/fundamentals/stellar-data-structures/accounts
+- Stellar Asset Contract: https://developers.stellar.org/docs/tokens/stellar-asset-contract
+- Soroban smart contracts overview: https://developers.stellar.org/docs/build/smart-contracts/overview
+
+## Reward Scenarios
+
+Potential reward triggers:
+
+- Complete a first study session.
+- Maintain a three-day or seven-day study streak.
+- Complete a quiz set with a minimum accuracy threshold.
+- Review a previously generated card after a delay.
+- Finish a topic review flow without skipping every card.
+- Use notes or PDF upload to create study material and then complete a review.
+
+Reward design should favor consistency and meaningful engagement over raw clicks. Streak rewards should not require users to disclose sensitive topics, upload private documents, or connect a wallet before they understand the tradeoff.
+
+## Stellar Asset Options
+
+### Off-Chain Points First
+
+Study Copilot can start with internal points that are not on-chain and cannot be transferred.
+
+Best fit:
+
+- Earliest MVP.
+- UX experiments.
+- Tuning streak rules before adding token value.
+
+Tradeoffs:
+
+- Not portable outside the app.
+- Requires backend storage before points persist across sessions.
+- Not independently verifiable.
+
+### Classic Stellar Issued Asset
+
+An issuer account can create a custom Stellar asset for rewards. Learners would need a Stellar account and trustline before receiving that asset.
+
+Best fit:
+
+- Simple transferable reward experiments.
+- Wallet-native balances.
+- Future community or sponsor reward programs.
+
+Tradeoffs:
+
+- Trustlines add UX complexity.
+- Transferable assets can create market expectations.
+- Issuer account custody and supply controls need operational planning.
+- Mainnet issuance raises compliance, tax, and support questions.
+
+### Stellar Asset Contract Or Soroban Token
+
+The Stellar Asset Contract and Soroban token patterns can represent assets in smart-contract flows, including richer rules around minting, distribution, and integration with other contracts.
+
+Best fit:
+
+- Future programmable rewards.
+- Badge-gated or contract-verified reward flows.
+- Testnet contract prototypes.
+
+Tradeoffs:
+
+- More implementation complexity than off-chain points.
+- Requires contract audits before real value is involved.
+- Still needs anti-abuse controls outside the contract.
+
+### Non-Transferable Credits
+
+Rewards can be modeled as non-transferable credits or achievement-linked balances rather than open tokens.
+
+Best fit:
+
+- Learning motivation without creating a tradable asset.
+- Lower economic risk.
+- Badge or achievement unlocks.
+
+Tradeoffs:
+
+- Requires custom rules or off-chain enforcement.
+- Less portable than normal Stellar assets.
+- Users may still perceive credits as monetary value.
+
+## Suggested Reward Flow
+
+1. Learner studies without a wallet requirement.
+2. App tracks local progress signals, such as cards viewed, quiz attempts, and streak days.
+3. Backend verifies the reward rule and checks rate limits.
+4. App shows a clear explanation of the earned reward.
+5. If using off-chain points, backend updates the learner profile.
+6. If using Stellar rewards, learner connects a verified wallet.
+7. Backend queues or submits the reward transaction only after abuse checks pass.
+8. App shows reward status: pending, issued, failed, or held for review.
+
+## Abuse Risks
+
+- Scripted clients can replay study sessions or quiz requests.
+- Users can create many accounts or wallets to farm first-session rewards.
+- Users can upload trivial or repeated content to trigger study flows.
+- Learners may brute-force quizzes if reward rules only check correctness.
+- Timezone and clock manipulation can fake streak boundaries.
+- Transferable assets can encourage farming, resale, or bot participation.
+- Reward incentives can reduce learning quality if points are easier than understanding.
+
+## Mitigation Ideas
+
+- Keep initial rewards off-chain and low/no monetary value.
+- Use server-side streak state instead of trusting client timestamps.
+- Rate-limit reward-eligible actions per learner, device, IP range, and wallet.
+- Require meaningful intervals between streak days.
+- Add minimum engagement checks, such as card dwell time plus quiz attempts.
+- Cap daily rewards and add cooldowns.
+- Delay on-chain issuance until a reward passes review thresholds.
+- Use testnet only while reward economics are being validated.
+- Pair rewards with anti-abuse research before enabling mainnet assets.
+
+## MVP Recommendation
+
+Rewards should be future work, not the first Stellar MVP.
+
+Recommended path:
+
+1. Ship wallet identity and achievement badges first.
+2. Add off-chain, non-transferable streak points for experimentation.
+3. Measure whether points improve study consistency without encouraging low-quality behavior.
+4. Add testnet Stellar reward prototypes only after anti-abuse controls exist.
+5. Defer mainnet or transferable rewards until legal, compliance, treasury, and support questions are answered.
+
+This keeps Study Copilot focused on learning while still leaving a path to Stellar-native incentives later.
+
+## Open Questions
+
+- What reward value, if any, is appropriate for study streaks?
+- Should rewards depend on quiz accuracy, completion, or verified time spent?
+- How should users recover rewards if they lose a wallet?
+- Should rewards be redeemable, transferable, or purely symbolic?
+- Who funds and controls reward issuance?
+- What user notice is needed before study behavior becomes reward-eligible?