prompt-engineering-expert

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted user data in the form of prompts submitted for evaluation.
  • Ingestion points: User-provided strings for analysis, documented in README.md, CLAUDE.md, and examples/EXAMPLES.md.
  • Boundary markers: The skill instructions recommend using XML tags and structured placeholders (e.g., [YOUR PROMPT]) to delimit user content.
  • Capability inventory: The skill defines its scope as "Analysis Only" and explicitly states it does not execute code or run prompts. It has no access to subprocess tools, network operations, or file-writing capabilities.
  • Sanitization: Input content is not explicitly sanitized, but the lack of exploitable tools effectively mitigates the risk of indirect injection.
  • [SAFE]: The skill provides high-quality educational content based on official documentation. All external references are to trusted or well-known sources, such as Anthropic's official blog and documentation, and reputable prompt engineering educational sites.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 01:51 PM
Security Audit — agent-trust-hub — prompt-engineering-expert