The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides command-line templates using 'curl' to interact with remote web servers for testing purposes. These commands allow the user to perform tasks like forced browsing, parameter tampering, and checking for insecure transport protocols.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes user-provided target information. Ingestion points: The 'TARGET' variable used as a placeholder in shell commands throughout the SKILL.md file. Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the target input as untrusted data. Capability inventory: The skill uses subprocess calls to 'curl' for network operations and 'python3' for payload generation. Sanitization: No sanitization or validation of the input target URL is implemented in the provided methodology.
[DYNAMIC_EXECUTION]: The skill includes a 'python3' command execution block used to generate a base64-encoded PHP serialization payload. This is used to demonstrate how to test for insecure deserialization vulnerabilities.

auth-bypass