Bundled with Unleash skills pack. Upstream: https://github.com/zhaoxuya520/reverse-skill
Database Security Assessment
ACTION REQUIRED (execute immediately after reading)
NOW: read precedent-pentest; destructive statements against production databases are forbidden unless explicitly allowedNOW: write scope clearly: instances, account permissions, whether writes/deletes are allowedNEXT: client tool pathsACT: exposure surface → authentication → authorization → configuration → (safe) exploit-chain verification
Applicable Scenarios
- Unauthenticated databases/weak passwords/wrongly bound 0.0.0.0
- Excessive privileges, dangerous features (xp_cmdshell, COPY PROGRAM, UDF)
- Lateral movement: from app account to DBA
- NoSQL injection and Redis file writes etc. (authorized environments)
Workflow
□ Network exposure and TLS
□ Account roles and grantees
□ Sensitive table access controls
□ Dangerous configuration: file_priv, xp_cmdshell, load_file
□ Whether audit logging is enabled
□ Backup and snapshot permissions
Toolchain
| Tool | Purpose |
|---|---|
| Official CLIs | Connection and enumeration |
| sqlmap | Injection verification (authorized) |
| nuclei | Known exposure templates |
| Cloud RDS console audit | Configuration |
References
references/db-misconfig-checklist.md../pentest-tools/../cloud-k8s/
Routing Context
Upstream: MASTER R35
Downstream: OS command obtained → attack-chain; cloud-managed → cloud-k8s
Task Completion Self-Check
- Avoided unauthorized writes/deletes?
- Distinguished configuration issues from exploitable chains?
- Checklist?