Skills
Navigate
skills/skills-shell/superpowers/prompt-engineering/Gen Agent Trust Hub

prompt-engineering

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides several bash examples for executing model runs via the belt CLI tool.
  • [EXTERNAL_DOWNLOADS]: The skill refers to external resources for installation and additional modules from the inference-sh GitHub repository.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides prompt templates that interpolate untrusted user data.
  • Ingestion points: SKILL.md examples use --input parameters for user content.
  • Boundary markers: No specific delimiters are used to wrap external content in the examples.
  • Capability inventory: Shell execution capability is provided through the belt CLI tool.
  • Sanitization: No sanitization of user-provided content is present in the demonstrated templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:23 PM
Security Audit — agent-trust-hub — prompt-engineering