T09 · Insecure Skill Coding Practices
Warning
- Location
scripts/extract_rides_gateway.py:98- Finding
Untrusted Email Content Can Manipulate Model-Extracted Ride Records
- Content
View full analysis
- Remediation
View remediation
Security audit
windylam
Security checks for vulnerabilities and agentic risk
Overview
This skill is mostly coherent and local-first, but it handles sensitive Gmail ride receipts through model extraction and CSV export without enough validation or output neutralization.
Install only if you are comfortable granting the skill access to ride receipt emails and storing raw receipt HTML locally. Keep the generated JSON, SQLite, and CSV files private, delete raw emails.json when finished, and be cautious opening or sharing the CSV until formula neutralization and stricter extraction validation are added.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
- Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
- Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
- Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)
T09 · Insecure Skill Coding Practices
Warning
- Location
scripts/export_anonymized_rides_csv.py:166- Finding
Model-Derived Values Are Exported Without Spreadsheet Formula Neutralization
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Behavioral ASTexec() Call, eval() Call, Dynamic Import
- Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)
Tainted flow: 'req' from os.environ.get (line 104, credential/environment) → urllib.request.urlopen (network output)
Critical
- Category
- Data Flow
- Confidence
- 90% confidence
- Finding
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
- Content
python }, method='POST', ) with urllib.request.urlopen(req, timeout=timeout) as resp: body = json.loads(resp.read().decode('utf-8')) for item in body.get('output', []): if item.get('type') == 'message':
subprocess module call
Medium
- Category
- Dangerous Code Execution
- Confidence
- 70% confidence
- Finding
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
- Content
python last_err = None for attempt in range(1, retries + 1): try: out = subprocess.check_output(cmd, text=True, stderr=subprocess.STDOUT) return json.loads(out) except subprocess.CalledProcessError as e: last_err = e
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The script fetches full Gmail message content, including HTML bodies and metadata, and writes it to a local JSON file without any consent prompt, warning, redaction, or minimization. Because ride receipts can contain sensitive personal and financial information, this creates a real privacy and data-handling risk if the output file is stored insecurely, shared, or processed by other tools.
- Content
Missing User Warnings
Low
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The code writes a CSV to a user-specified path and creates parent directories if needed, which is a file-system side effect involving potentially sensitive ride data. Although the filename and docstring mention export, there is no explicit confirmation prompt or user-facing warning in the code about writing data to disk.
- Content
Static analysis
No suspicious patterns detected.