mthds-explain

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to verify the local environment and manage its toolchain.
  • Evidence: Step 0 runs a bash script to identify and execute the latest version of the mthds-env-check utility from the local plugin cache.
  • Evidence: Steps 6 and 7 utilize the mthds-agent CLI for bundle validation and execution.
  • [EXTERNAL_DOWNLOADS]: The skill directs users to install necessary vendor tools.
  • Evidence: Prompting the user to run npm install -g mthds if the agent is not found or is outdated.
  • [DATA_EXFILTRATION]: The skill reads and analyzes external files, which creates a surface for indirect prompt injection.
  • Ingestion points: The skill reads the entire contents of .mthds bundle files in Step 1.
  • Boundary markers: Absent; there are no explicit instructions to the agent to ignore instructions embedded within the analyzed data.
  • Capability inventory: The agent has access to Bash, Read, Write, and Edit tools which could be targeted by malicious data in the bundle.
  • Sanitization: Absent; data is analyzed and summarized directly without escaping or validation for embedded commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 01:46 AM
Security Audit — agent-trust-hub — mthds-explain