skills/maragudk/fabrik/observability/Gen Agent Trust Hub

observability

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates an environment where an AI agent processes telemetry data (logs, traces, events) which can originate from untrusted external sources, creating an indirect injection surface.
  • Ingestion points: Production telemetry and wide event attributes as described in SKILL.md and references/wide-event-attributes.md.
  • Boundary markers: The skill does not define explicit delimiters for telemetry data to prevent the agent from interpreting embedded instructions as commands.
  • Capability inventory: The skill is intended for use by agents that 'query telemetry, call APIs, [and] run code', as described in references/ai-llm-and-agents.md.
  • Sanitization: The documentation recommends PII redaction but does not specify methods for sanitizing natural language instructions within structured log fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 02:00 AM
Security Audit — agent-trust-hub — observability