The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides command-line templates for security testing tools such as Hydra and Burp Suite. These are intended for authorized testing of authentication endpoints (e.g., brute-forcing credentials, testing rate limiting).
[EXTERNAL_DOWNLOADS]: The skill mentions the use of common security wordlists like 'rockyou.txt', which is a standard resource in penetration testing environments.
[DYNAMIC_EXECUTION]: Phase 6 includes a Python script template that uses the requests and hashlib libraries to collect and analyze session tokens for entropy and patterns.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze responses from external web applications (Phase 3 and Phase 9). This introduces a surface where untrusted data from a target server could potentially influence the agent's behavior, although no specific automated parsing of complex untrusted content is present.
Ingestion points: Server responses from login, registration, and password reset endpoints (SKILL.md Phase 1, 3, 9).
Boundary markers: None provided in the instruction templates.
Capability inventory: Network requests via Python requests and tool execution via Hydra.
Sanitization: The skill focuses on manual analysis and standard tool output interpretation; explicit data sanitization instructions for the agent are not defined.

Broken Authentication Testing