sentry-automation
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. It retrieves sensitive error data, stack traces, and issue details from Sentry, which can contain instructions injected by external users triggering application errors. If an agent triaging these issues follows embedded instructions, it could perform unauthorized actions using the skill's capabilities.
- Ingestion points:
SENTRY_GET_ORGANIZATION_ISSUE_DETAILS,SENTRY_RETRIEVE_AN_ISSUE_EVENT, andSENTRY_LIST_AN_ISSUES_EVENTSinSKILL.md. - Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent about ignoring potential commands within Sentry data.
- Capability inventory: The skill includes several administrative tools such as
SENTRY_CREATE_PROJECT_RULE_FOR_ALERTS,SENTRY_UPDATE_A_MONITOR, andSENTRY_CREATE_RELEASE_FOR_ORGANIZATIONinSKILL.md. - Sanitization: Absent. There is no evidence of validation or sanitization for the data retrieved from Sentry before it is introduced into the agent's context.
Audit Metadata