pii-checker

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the agent-sec-cli command-line tool to perform PII scanning. The instructions provide explicit security requirements to use parameter arrays or robust shell quoting (e.g., shlex.quote) to mitigate command injection risks when handling file paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data (logs, files, and arbitrary text) to detect sensitive information, creating a potential surface for indirect prompt injection.
  • Ingestion points: External text files (via the --input flag), standard input (via --stdin), and raw text strings passed to the agent.
  • Boundary markers: No specific boundary delimiters are defined in the instructions for isolating the content being scanned.
  • Capability inventory: The skill has the ability to execute subprocesses (agent-sec-cli), read local files, and write redacted output to the file system upon user request.
  • Sanitization: The skill instructions specifically advocate for using standard input or file paths rather than command-line arguments to handle sensitive content, which prevents data leakage into process lists and mitigates shell-based concatenation attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:58 AM