signoz-creating-alerts
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill does not contain any malicious patterns, obfuscation, or unauthorized data exfiltration techniques. It adheres to vendor-specific best practices for managing SigNoz monitoring infrastructure and implements mandatory verification steps before modifying system state.
- [CREDENTIALS_SAFE]: The skill provides explicit instructions to prevent the accidental exposure of sensitive data. Step 7 ('Never echo channel secrets') mandates that secrets provided by the user, such as Slack webhook URLs or PagerDuty integration keys, must never be echoed back in chat, stashed in persistent clarification contexts, or included in any visible summaries or navigation suggestions.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a natural ingestion surface for untrusted data as it processes user-provided alert intents and external resource metadata to generate configurations. This is inherent to the skill's primary purpose.
- Ingestion points: User-provided alert intent ($ARGUMENTS) and resource attributes discovered via SigNoz tools (e.g., service.name, host.name, k8s.namespace.name).
- Boundary markers: The skill builds structured JSON payloads (schemaVersion v2alpha1) for API interactions, providing inherent separation between instructions and data.
- Capability inventory: The skill uses
signoz_create_alertandsignoz_create_notification_channelto modify system state, andsignoz_execute_builder_queryto perform data validation and dry-runs. - Sanitization: The skill follows strict schema validation (Step 5) and formatting rules, such as mapping signal phrasing to specific alert types and using OpenTelemetry semantic conventions for attribute names.
Audit Metadata