senior-prompt-engineer

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Comprehensive analysis of the provided Python scripts and documentation shows no evidence of malicious intent. The skill focuses on legitimate prompt engineering tasks, and its scripts (prompt_optimizer.py, rag_evaluator.py, agent_orchestrator.py) use standard libraries for local text processing and data evaluation. No external network requests, unauthorized file access, or obfuscated code segments were identified.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process external prompt text and datasets for analysis.
  • Ingestion points: scripts/prompt_optimizer.py (prompt files), scripts/rag_evaluator.py (JSON question/context files), and scripts/agent_orchestrator.py (agent configurations).
  • Boundary markers: The scripts do not implement specific delimiters or 'ignore-previous-instructions' guardrails when parsing the input files.
  • Capability inventory: The tools are limited to text analysis, metric calculation, and report generation; they lack dangerous capabilities such as shell execution, network communication, or administrative system access.
  • Sanitization: Input data is handled as raw text or structured objects without explicit sanitization against embedded instruction overrides, which is consistent with the skill's purpose as a developer-facing analysis toolkit.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 05:54 AM