The Agent Skills Directory

[SAFE]: The skill serves as a legitimate security research resource, providing methodologies for identifying and validating vulnerabilities in ASP.NET Webforms, SharePoint, and WCF services.
[COMMAND_EXECUTION]: The skill provides curl command examples and a Python script for network-based security probing. These tools are designed to interact with remote target servers provided by the user and do not contain logic to execute malicious commands on the host environment.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with external web server responses (headers and HTML bodies), creating a potential ingestion surface for untrusted data.
Ingestion points: HTTP response headers (e.g., X-AspNet-Version) and form fields (e.g., __VIEWSTATE) from remote targets in SKILL.md.
Boundary markers: The skill employs targeted regex patterns (e.g., re.search(r'__VIEWSTATE" id="__VIEWSTATE" value="([^"]+)', r.text)) to isolate specific data fields, preventing the ingestion of arbitrary response content.
Capability inventory: The skill is limited to making network requests (requests.get, requests.post) and performing string analysis. It lacks the capability to write files, modify system settings, or execute shell commands based on the ingested data.
Sanitization: Data is processed strictly as text for pattern matching and status code verification; there is no unsafe interpolation into executable contexts.

hunt-aspnet