The Agent Skills Directory

[COMMAND_EXECUTION]: The script uses subprocess.run to call grep or ripgrep for identifying code origin. This is implemented using argument lists rather than shell strings, which effectively prevents command injection from untrusted API data.
[EXTERNAL_DOWNLOADS]: The script performs network requests to the Brave Backtrace API and the Brave Release Schedule wiki on GitHub. These operations are restricted to the HTTPS protocol and target trusted vendor-controlled endpoints.
[DATA_EXFILTRATION]: While the skill accesses crash data, it implements PII-safe practices by sanitizing file paths in stack traces and only outputting aggregate data. Credentials are managed safely through environment variables.

top-crashers