database-attack
Fail
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: CRITICALCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONPERSISTENCEOBFUSCATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute a variety of security auditing and database administration tools, including odat, oscanner, sqsh, redis-cli, mongoaudit, patator, ncrack, and hydra, as shown in the methodologies of SKILL.md and multiple guide files.
- [REMOTE_CODE_EXECUTION]: The skill provides extensive payloads and methodologies for achieving code execution on remote database targets. Examples include enabling and using xp_cmdshell in MSSQL (guides/database-lateral-movement-guide.md), loading User Defined Functions (UDF) in MySQL, abusing COPY TO PROGRAM in PostgreSQL, and leveraging DBMS_SCHEDULER in Oracle (payloads.md).
- [DATA_EXFILTRATION]: Payloads are included for exfiltrating data from target databases using native features (e.g., MySQL LOAD_FILE, PostgreSQL COPY) and network-based methods like DNS exfiltration via MSSQL xp_cmdshell and HTTP exfiltration from MySQL (payloads.md).
- [PRIVILEGE_ESCALATION]: The skill documents paths for escalating access within databases, such as exploiting GRANT OPTION in MySQL, superuser roles in PostgreSQL, and database links in Oracle (guides/database-lateral-movement-guide.md).
- [PERSISTENCE]: Instructional guides demonstrate how to maintain access on compromised systems by injecting SSH authorized_keys or writing web shells and cron jobs through misconfigured Redis instances (guides/nosql-attack-guide.md).
- [OBFUSCATION]: A PowerShell reverse shell payload in 'guides/database-lateral-movement-guide.md' is presented in Base64-encoded format ('powershell -e JABjAGwAaQBlAG4AdAA...'). While this is an obfuscation technique, it is used here for educational demonstration of target exploitation.
- [INDIRECT_PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to ingest and process data from external targets (e.g., database query results, service banners). Ingestion points are located in CLI arguments across scripts like 'guides/nosql-attack-guide.md' and 'guides/database-bruteforce.md'. The skill uses tools like Bash and Read but lacks explicit sanitization or boundary markers for these external data streams.
Recommendations
- Contains 4 malicious URL(s) - DO NOT USE
Audit Metadata