The Agent Skills Directory

[SAFE]: The skill and its associated reference documentation provide legitimate regulatory guidance for medical device classification and technical file preparation without any detected security risks.
[COMMAND_EXECUTION]: The provided Python script scripts/mdr_gap_analyzer.py is a safe local utility for assessing compliance gaps. It relies solely on Python standard libraries and performs no network operations, privilege escalation, or access to sensitive local files.
[DATA_EXFILTRATION]: There is no evidence of unauthorized data access or transmission. The skill does not request or handle sensitive credentials, environment variables, or private user data.
[PROMPT_INJECTION]: The instructions are task-oriented and follow professional guidelines, with no patterns identified that attempt to bypass AI safety filters, override system instructions, or extract system prompts.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests user-provided device descriptions for analysis, it lacks the necessary capabilities (such as network access or arbitrary file writing) to facilitate an injection-based attack. The local analysis script is restricted to pre-defined compliance checks.

mdr-745-specialist