The Agent Skills Directory

[DATA_EXPOSURE]: The skill handles SENTRY_AUTH_TOKEN securely by instructing the user to set it as an environment variable rather than sharing it in the chat interface. It also includes a redact_data function in the core script that automatically scrubs potential PII, such as email addresses and IP addresses, from API responses before they are presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to sentry.io, which is a well-known and trusted observability service. No third-party or untrusted domains are contacted.
[COMMAND_EXECUTION]: The skill executes a bundled Python script (scripts/sentry_api.py) for API interactions. This script uses only standard library modules and performs specific, predefined read-only tasks.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external Sentry issue reports (e.g., issue titles or descriptions) which could contain malicious instructions.
Ingestion points: Data enters the context via the scripts/sentry_api.py script which fetches issue details and events from the Sentry API.
Boundary markers: The instructions do not define clear delimiters or warnings for the agent to ignore instructions embedded in the external data.
Capability inventory: The skill can perform network GET requests to Sentry; it has no file-write or arbitrary code execution capabilities.
Sanitization: The script performs PII redaction (email/IP), but does not sanitize the content for malicious prompt instructions.

sentry