Bounties are automatically suggested and maintained by the AI agent.
Each entry maps to an issue or PR with a corresponding token reward.
| ID | Issue/PR | Difficulty | Suggested Reward (OSS Tokens) | Status |
|---|---|---|---|---|
| 1 | #12 Add CLI interface | 🟠 Medium | 200 | 🟢 Open |
| 2 | #18 Improve AI prompt logic | 🔵 Hard | 350 | 🟢 Open |
| 3 | #25 Add test coverage | 🟢 Easy | 100 | 🟢 Open |
The AI Maintainer evaluates and dynamically adjusts bounty rewards using multiple signals, including:
| Criterion | Description |
|---|---|
| Complexity | Number of files and lines changed, technical depth of logic, and potential side effects. |
| Impact | Alignment with core project goals and how much it improves usability, performance, or maintainability. |
| Clarity & Quality | Code readability, comments, documentation added, and adherence to contribution guidelines. |
| Timeliness | How quickly the issue or PR was addressed relative to its creation date and urgency level. |
| Issue Age & Priority | Older or high-priority issues may receive a bonus multiplier to incentivize completion. |
| Review Feedback | AI analyzes peer or maintainer reviews to assess collaboration and improvement from feedback. |
Each factor contributes to a composite bounty score, which determines the suggested reward tier:
| Difficulty | Reward Range (Tokens) |
|---|---|
| 🟢 Easy | 50–150 |
| 🟠 Medium | 150–300 |
| 🔵 Hard | 300–500 |
| 🔴 Priority / Legacy Issue | 500–800 |
open, in_progress, completed, Ready for reward, Rewarded