The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze log.txt, which contains captured network traffic from mitmproxy. Since network traffic originates from external, potentially untrusted sources, it serves as an attack surface for indirect prompt injection. An attacker could craft specific HTTP responses or requests that, when logged and subsequently analyzed by the agent, might attempt to influence the agent's reasoning or output.\n
Ingestion points: The file log.txt in the current working directory.\n
Boundary markers: No explicit markers or delimiters are defined to separate untrusted log data from the agent's instructions.\n
Capability inventory: The skill uses shell commands including curl, grep, and mitmdump.\n
Sanitization: There are no instructions provided to sanitize or validate the content of the logs before analysis.- [COMMAND_EXECUTION]: The skill provides templates for shell commands and loops (using curl and for loops) to perform security testing against target endpoints. While these are intended for legitimate security auditing, they involve executing commands based on user-provided arguments ($ARGUMENTS).

mitm-find-otp