exploring-apm-traces

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes trace data from PostHog APM, which could potentially contain malicious instructions embedded in span names or attributes. \n
  • Ingestion points: Data is fetched via posthog:apm-trace-get and posthog:query-apm-spans MCP tools, which is then parsed by scripts in the scripts/ directory. \n
  • Boundary markers: The skill does not explicitly use boundary markers or instructions to disregard embedded commands in the data. \n
  • Capability inventory: The skill utilizes Python scripts to process and display trace data within the agent's context. \n
  • Sanitization: Data values are truncated but not sanitized for instruction-like strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:42 AM
Security Audit — agent-trust-hub — exploring-apm-traces