skills/openclaw/skills/truthcheck/Gen Agent Trust Hub

truthcheck

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'truthcheck' Python package. This is a vendor-provided tool used for the skill's primary fact-checking functionality.\n- [COMMAND_EXECUTION]: The skill performs shell commands using the 'truthcheck' binary to verify claims, check URL reputation, and trace information origins.\n- [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by passing user-supplied text directly to CLI commands.\n
  • Ingestion points: User input is processed as arguments for the verify, check, and trace subcommands in SKILL.md.\n
  • Boundary markers: The documentation includes examples that wrap claim text in double quotes, providing basic isolation, though it remains vulnerable to escaped or adversarial input.\n
  • Capability inventory: The CLI tool can execute network searches and read local files (via the -f flag), which are powerful capabilities if combined with successful injection.\n
  • Sanitization: There is no explicit mention of input validation or sanitization within the skill's instructions to prevent command-line injection or prompt manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 01:20 PM