skills/comet-ml/opik-skills/agent-ops/Gen Agent Trust Hub

agent-ops

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for using the opik connect CLI tool to run local agent scripts (e.g., python3 app.py or npx tsx app.ts). This is used for pairing local development environments with the Opik UI for tracing and monitoring.
  • [EXTERNAL_DOWNLOADS]: The skill references the opik Python library and CLI, as well as the tsx runner for TypeScript. These are official tools and standard development dependencies provided by the author (comet-ml).
  • [PROMPT_INJECTION]: The skill addresses indirect prompt injection risks by providing specific code examples for input validation and sanitization. It demonstrates how to filter common injection patterns like 'ignore previous' from retrieved external content and suggests the use of guardrails to enforce safety boundaries.
  • [DATA_EXFILTRATION]: The skill emphasizes data privacy by providing documentation on PII masking using anonymizers and selective logging to prevent sensitive information from being recorded in traces.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 11:49 AM