The Agent Skills Directory

[SAFE]: No security issues detected across all analyzed categories.
[PROMPT_INJECTION]: The instructions provide a clear framework for report generation and evaluation. There are no attempts to bypass safety guidelines, override system instructions, or extract system prompts.
[DATA_EXFILTRATION]: The skill does not perform network requests, access sensitive local files (like SSH keys or environment variables), or handle hardcoded credentials.
[REMOTE_CODE_EXECUTION]: No remote scripts, package installations, or dynamic code execution patterns (like eval or subprocess calls) were found.
[INDIRECT_PROMPT_INJECTION]: While the skill processes influencer data (handles and public content), it uses this information solely for structured scoring and reporting. There is no evidence of unsafe interpolation of untrusted data into executable contexts.
[OBFUSCATION]: All content is provided in plain text. No Base64, hex encoding, zero-width characters, or other concealment techniques were detected.

fit-scorer