Type: Directory Traversal
Product: DB-GPT
Affected versions: 0.8.0
Affected component: packages/dbgpt-app/src/dbgpt_app/openapi/api_v1/agentic_data_api.py:skill_upload
Suggested CWE: CWE-22 (Directory Traversal)
Reference: eosphoros-ai/DB-GPT#3026
DB-GPT 0.8.0 contains directory traversal in skill_upload (packages/dbgpt-app/src/dbgpt_app/openapi/api_v1/agentic_data_api.py:40). A remote attacker can use the validated exploitation path to write files outside the intended workspace or storage boundary.
Condensed from the authorized local validation used for the original report; local filesystem paths are replaced with <local-path>.
=== STEP 1: Setup Environment ===
- Clone repository: git clone https://github.com/eosphoros-ai/DB-GPT.git
- Checkout vulnerable version: git checkout v0.8.0
- Navigate to project root: cd DB-GPT
- Install dependencies: pip install -e ".[default]"
- Configure environment: cp .env.template .env
- Start DB-GPT service: python -m dbgpt.app.dbgpt_server === STEP 2: Verify Service Running ===
- Wait for: "Application startup complete" message
- Service runs on: http://localhost:5670
- Verify endpoint exists: curl http://localhost:5670/api/v1/skills/upload === STEP 3: Create Malicious Skill File === Create a test file (malicious_skill.py) with path traversal filename: echo "print('Malicious code executed')" > test_payload.py …
Ro1ME