The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown-based technical reference papers and procedural instructions aimed at improving the security of identity and access management systems. No executable code or suspicious scripting was found across all 14 files.
[NO_CODE]: The skill does not implement any custom tools, scripts, or binary executables. It functions primarily as an instruction-tuning extension for an AI agent, providing it with normative security requirements derived from industry standards like NIST and OWASP.
[EXTERNAL_DOWNLOADS]: The skill contains numerous URLs referencing well-known and trusted standards organizations, including IETF (rfc-editor.org), NIST (nist.gov), OWASP (owasp.org), and W3C (w3.org). These are used solely for documentation and verification purposes.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied requests for authentication and authorization design, it implements strong mitigation strategies. The mandatory workflow requires the agent to consult specific reference papers, answer pre-implementation security questions, and adhere to a 'Stop Condition' list that prevents insecure configurations like plaintext secrets or account enumeration.

auth-access