web-access-control

Fail

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for using and installing several external security tools, including ffuf, sqlmap, and Burp Suite. The included Python scripts require the installation of third-party libraries such as requests and PyJWT to perform network operations and JWT manipulation.
  • [COMMAND_EXECUTION]: Multiple markdown files and scripts contain command-line templates for Bash and Python, enabling the automated enumeration of target web applications. These include curl-based IDOR scanners and path traversal probes.
  • [DYNAMIC_EXECUTION]: The skill includes a robust automated scanner (access_control_scanner.py) that dynamically generates and executes HTTP requests based on user-provided base URLs and authentication tokens. It also performs runtime calculations for ID ranges and payload construction.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted responses from external web servers during its scanning operations.
  • Ingestion points: The scripts in the guides/ directory ingest raw HTML and JSON responses from target URLs provided by the user.
  • Boundary markers: The scripts lack explicit boundary markers or delimiters for ingested content, instead relying on regular expression matching to identify vulnerability indicators.
  • Capability inventory: The skill possesses network access capabilities through the requests library and local file-write capabilities for generating Markdown and JSON reports.
  • Sanitization: The scripts search for specific strings (e.g., 'root:', 'password', 'ssn') to detect vulnerabilities but do not sanitize or encode the overall response data before processing or logging.
Recommendations
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Sep 19, 2026, 04:36 PM