The Agent Skills Directory

[Prompt Injection] (SAFE): The 'Angry Consultant' persona is used for creative output styling (aggressive cuts, specific typography) rather than as a mechanism to bypass AI safety guardrails or ignore system instructions.- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, environment variables, or hardcoded credentials were found. There are no network-capable commands for data exfiltration.- [Remote Code Execution] (SAFE): The file does not contain any commands to download or execute external scripts or binaries.- [Indirect Prompt Injection] (INFO): The skill identifies a 'Retention Audit' capability which suggests it may process external content. However, in its current form, it only provides a description and lacks the implementation code where injection vulnerabilities typically occur.- [No Code] (INFO): The analyzed file is strictly a markdown definition (SKILL.md). No accompanying scripts or tool implementations were provided for logic-level security analysis.

angel-core