find-leads

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines a workflow that executes a bash command involving the orth CLI tool and a piped Python one-liner. This is used to fetch profile data and parse JSON results into a markdown table.
  • [PROMPT_INJECTION]: The skill instructions direct the agent to replace a placeholder (REPLACE_WITH_QUERY) in a shell command with raw user input. This creates a vulnerability surface for indirect prompt injection or command injection if a user provides a query designed to break out of the shell quotes or JSON structure.
  • Ingestion points: User-provided search query interpolated into the workflow command in SKILL.md.
  • Boundary markers: The input is placed inside a JSON string value which is itself inside a single-quoted bash argument.
  • Capability inventory: The skill uses bash and python3 subprocess execution.
  • Sanitization: No explicit sanitization or escaping logic is provided in the skill script; it relies on the underlying platform's input handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 05:45 AM
Security Audit — agent-trust-hub — find-leads