sentry-fix-issues
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill proactively addresses indirect prompt injection from Sentry event data (error messages, breadcrumbs) by instructing the agent to treat it as untrusted input and ignore any directive-like content. 1. Ingestion points: Sentry MCP tools such as get_issue_details and search_issue_events. 2. Boundary markers: The skill includes a dedicated 'Security Constraints' section defining the trust boundary. 3. Capability inventory: The skill has the capability to read repository files and suggest or apply code fixes. 4. Sanitization: The skill mandates redaction of PII and secrets, and requires verification of Sentry data against the local codebase before taking action.
- [DATA_EXFILTRATION]: Security rules are in place to prevent the reproduction of sensitive tokens, passwords, or PII found in Sentry logs in any output generated by the agent.
Audit Metadata