run-prompt

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by reading content from local files and executing it as agent instructions.
  • Ingestion points: The skill reads files matching the pattern ./prompts/[number]-*.md.
  • Boundary markers: Absent. The instructions explicitly state to 'Execute the prompt content as instructions' without delimiters or warnings to ignore embedded directives.
  • Capability inventory: The skill uses file system read/write operations and shell command execution via the agent's environment.
  • Sanitization: No validation, escaping, or filtering is performed on the content of the prompt files before execution.
  • [COMMAND_EXECUTION]: The skill uses a shell command string (ls ./prompts/ 2>/dev/null | grep -E "^[0-9]{3}-" | sort -V) to discover and organize the prompt files available for execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:39 AM
Security Audit — agent-trust-hub — run-prompt