The Agent Skills Directory

[SAFE]: The skill is designed for read-only access to Sentry data. It correctly implements authentication using the SENTRY_AUTH_TOKEN environment variable and explicitly warns users against pasting tokens into the chat, following secure credential management practices.
[SAFE]: The bundled Python script (scripts/sentry_api.py) includes a redaction mechanism that recursively scans API responses to mask potential PII, specifically targeting email addresses and IP addresses using regular expressions and key-name matching (e.g., 'email', 'ip_address').
[SAFE]: Network operations are restricted to communication with Sentry's official API (sentry.io). As a well-known service used for the skill's primary purpose, this behavior is considered safe and standard for observability tools.
[SAFE]: All dependencies used in the script are part of the Python standard library (argparse, json, os, re, urllib), ensuring no risk from external or unpinned third-party packages.

sentry