The Agent Skills Directory

[SAFE]: The skill defines a procedural workflow for triaging security vulnerabilities using a "7-Question Gate" and "Pre-Submission Gates." These are defensive heuristics designed to improve the accuracy of security analysis.
[COMMAND_EXECUTION]: Technical examples using curl are provided to demonstrate the difference between input validation and authentication layers. These commands use a placeholder domain (https://target) and are for educational and diagnostic purposes within an authorized context.
[PROMPT_INJECTION]: The skill does not contain instructions that attempt to bypass safety filters, jailbreak the agent, or override system guidelines. The "STOP" instructions mentioned in the triage process refer to the logical flow of finding validation, not the execution state of the AI agent.
[EXTERNAL_DOWNLOADS]: The skill references documentation and source code on GitHub for attribution and context. It does not contain commands to download, install, or execute external scripts or packages.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths (e.g., SSH keys, environment files), or exfiltration patterns were identified.

triage-validation