The Agent Skills Directory

[SAFE]: The skill consists of Markdown-based prompt templates and documentation. No executable scripts, binaries, or automated shell commands were found in the skill package.
[PROMPT_INJECTION]: No malicious prompt injection or jailbreak attempts were detected. The templates use instructional language to guide the agent in performing legitimate security tasks, such as STRIDE analysis and RBAC implementation, without attempting to bypass safety guardrails.
[DATA_EXFILTRATION]: No data exfiltration patterns or network-based exfiltration commands were found. The skill does not attempt to access sensitive files like ~/.ssh or ~/.aws, and placeholders are used for sensitive identifiers in documentation examples.
[OBFUSCATION]: No evidence of obfuscation techniques, such as Base64 encoding of commands, zero-width characters, or homoglyph attacks, was found in the text or metadata.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not utilize the dynamic context injection syntax (!command) to execute shell commands at load time.
[COMMAND_EXECUTION]: While the skill contains example curl commands for manual testing of webhooks and API endpoints, these are presented as documentation for the user and are not executed by the skill itself.

security-prompts