dast-scanning
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads official security testing resources from well-known sources, including the OWASP ZAP Docker image from the GitHub Container Registry and the Nuclei scanner from Project Discovery's official repository.
- [COMMAND_EXECUTION]: It provides various commands to run automated security scans, including Docker-based execution of ZAP scripts, installation of Nikto via system package managers, and execution of Nuclei templates.
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external web applications and APIs during scanning operations. This creates a theoretical surface where malicious application content could be ingested into the agent's context, although this risk is inherent to the primary purpose of security scanning.
- Ingestion points: Web responses and API specifications from target URLs provided by the user (SKILL.md, references/dast-tools.md).
- Boundary markers: The skill relies on standard tool output formats and does not include specific boundary delimiters for untrusted target data.
- Capability inventory: The skill utilizes network access for scanning, local file writing for reports, and subprocess execution for tool invocation.
- Sanitization: Tool outputs are generally processed as structured reports rather than direct prompt content.
Audit Metadata