signoz-setting-up-observability
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted telemetry data which represents a potential surface for indirect prompt injection attacks.
- Ingestion points: Data enters the agent context via
signoz_list_services,signoz_get_field_keys, andsignoz_get_field_valuesas described in Phase 3 ofSKILL.md. - Boundary markers: The skill instructs the agent to verify values and filter out common scanner paths (e.g.,
/.env,/.git), providing data-level boundaries but no explicit LLM-instruction delimiters to ignore commands inside telemetry strings. - Capability inventory: The skill utilizes tools to create production-impacting resources including
signoz_create_dashboard,signoz_create_alert, andsignoz_create_view. - Sanitization: While the skill enforces cardinality limits and mandatory dry-runs, it does not explicitly describe sanitization logic to filter LLM-oriented instructions from ingested telemetry data.
Audit Metadata