The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous shell command templates intended for execution by the agent or user. These include common security auditing tools such as subfinder, amass, nuclei, and subjack, as well as standard utilities like curl, dig, and bash loops for DNS resolution and fingerprinting.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection because it instructs the agent to fetch and process content from external, untrusted websites (target subdomains) to identify vulnerabilities.
Ingestion points: HTTP response bodies and headers retrieved from scanned subdomains during the manual verification and fingerprinting phases (found in SKILL.md).
Boundary markers: The provided scripts do not include specific delimiters or instructions to ignore potential commands embedded in the target's response content.
Capability inventory: The agent is granted extensive capabilities, including shell execution (bash), network requests (curl, dig), and file system operations (-o resolved.txt).
Sanitization: No sanitization or validation of the external content is implemented; the raw output of network requests is directly piped to text processing tools or presented to the agent's context.

hunt-subdomain