Skills
skills/davidcastagnetoa/skills/trace_id_propagation/Gen Agent Trust Hub

trace_id_propagation

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill follows security best practices for distributed tracing and does not contain any malicious patterns or unauthorized access attempts.\n- [COMMAND_EXECUTION]: The skill includes Python code to perform HTTP POST requests via the httpx library in SKILL.md. These requests are used to propagate tracing headers between internal microservices.\n- [PROMPT_INJECTION]: The skill processes untrusted header data in SKILL.md (ingestion point) and uses httpx for downstream requests (capability). It mitigates potential indirect injection by explicitly instructing the validation and sanitization of the traceparent format in Instruction 7.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 05:29 PM