The Agent Skills Directory

[PROMPT_INJECTION]: The skill establishes a clear persona and operational workflow for academic research coordination. No patterns were found that attempt to bypass AI safety guardrails, extract system prompts, or override core instructions.
[DATA_EXFILTRATION]: No unauthorized data transfer or credential exposure was detected. The skill manages research artifacts (matrices, drafts, reports) within a local project directory structure and does not contain network operations to external untrusted domains.
[REMOTE_CODE_EXECUTION]: There are no commands for downloading external scripts, installing unverified packages, or executing remote code. The collaboration relies on platform-native agent mentions rather than shell execution.
[COMMAND_EXECUTION]: No dangerous shell commands, privilege escalation (e.g., sudo), or persistence mechanisms (e.g., crontab) are present.
[SAFE]: The skill's behavior aligns perfectly with its metadata and stated purpose as a research supervisor tool. No obfuscation, metadata poisoning, or time-delayed triggers were identified.

survey-director