The Agent Skills Directory

[SAFE]: The skill implements a logical research framework using standard agent tools like web_search and financial_rigor. It does not contain any malicious instructions or unexpected behaviors.
[DATA_EXPOSURE]: No hardcoded credentials, secrets, or sensitive file paths (e.g., SSH keys, cloud configurations) were identified.
[REMOTE_CODE_EXECUTION]: There is no evidence of remote script downloading, execution of external payloads, or use of unsafe functions like eval() or exec().
[COMMAND_EXECUTION]: All tool calls are for intended research purposes. The skill does not attempt to execute arbitrary shell commands or escalate privileges.
[DATA_EXFILTRATION]: No network exfiltration patterns were found. File operations are restricted to writing research reports in a local reports directory.

bottleneck-hunter