The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes numerous shell command templates for using the ffuf utility. These templates allow the agent to perform automated network probing, such as directory discovery and parameter fuzzing, by executing a network-interacting tool through the system shell.
[INDIRECT_PROMPT_INJECTION]: The skill defines a vulnerability surface for indirect prompt injection by providing command-line templates that incorporate external data such as target URLs and wordlist paths.
Ingestion points: The skill expects the agent to populate command templates with target URLs and file paths provided by users or discovered in the execution environment (SKILL.md).
Boundary markers: There are no explicit instructions or delimiters provided to guide the agent in sanitizing or isolating these external inputs to prevent the execution of arbitrary shell commands.
Capability inventory: The skill enables the agent to utilize ffuf for broad network discovery and fuzzing, which involves generating arbitrary HTTP requests.
Sanitization: The skill does not include validation or sanitization logic to ensure that user-provided strings do not contain shell metacharacters or malicious payloads before being executed.

ffuf-fuzz