The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze web traffic logs (log.txt), which represent untrusted data from external network interactions. This creates a surface for potential indirect prompt injection if the logs were to contain malicious instructions.
Ingestion points: The skill reads and processes the log.txt file using search tools.
Boundary markers: The instructions do not define explicit delimiters or 'ignore' commands to isolate the log content from the agent's core instructions.
Capability inventory: The skill utilizes mitmdump, grep, sort, and curl to process the data and interact with external endpoints.
Sanitization: There is no evidence of sanitization or filtering applied to the log data before the agent processes it.
[COMMAND_EXECUTION]: The skill relies on shell commands to perform its vulnerability analysis functions. These tools are standard for the security research context intended by the skill.
Evidence: The skill provides examples of using mitmdump for traffic capture, grep for pattern identification in files, and curl for performing HTTP requests to verify authorization controls on target systems.

mitm-find-idor