mammumammi/Secure Soroban Attack Simulator
Auto-detects Soroban smart contracts in your repo, deploys each to testnet, simulates real attack vectors (authorization bypass, unauthorized drain, integer overflow, reentrancy) plus static analysis, and — optionally — runs a local AI agent (qwen2.5-coder via Ollama) that autonomously probes for vulnerabilities. Produces a pass/fail security report.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| network | Stellar network to run the simulation against | no | testnet |
| contracts_glob | Root path to scan for Soroban contracts | no | . |
| enable_ai_agent | Run the local qwen2.5-coder AI agent scan | no | true |
| fail_on_severity | Minimum severity that fails the action (low|medium|high|critical) | no | critical |
| ollama_model | Ollama model used by the AI agent | no | qwen2.5-coder:7b |
Outputs
| name | description |
|---|---|
| safe | true if no contract triggered a block at or above fail_on_severity |
| report_path | Path to the combined JSON report artifact |