markdownlint

Fail

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes instructions to suggest or execute 'sudo npm install -g markdownlint-cli' for remediation. Using 'sudo' allows for privilege escalation to root or administrator level on the host system.\n- [EXTERNAL_DOWNLOADS]: The skill installs the 'markdownlint-cli' package from the npm registry at runtime if it is missing from the system.\n- [COMMAND_EXECUTION]: The skill executes shell commands using file and directory paths provided in the agent's context (e.g., 'markdownlint '). This creates a potential for command injection if malicious file names or paths are processed.\n- [PROMPT_INJECTION]: The skill processes markdown files from the project environment which may contain malicious instructions designed to influence the agent's behavior during validation or reporting.\n
  • Ingestion points: Markdown files and directories (SKILL.md)\n
  • Boundary markers: Absent; files are passed directly to the linting tool.\n
  • Capability inventory: Shell command execution (npm, markdownlint) and file system writes (creating .markdownlint.json).\n
  • Sanitization: Absent; file paths and contents are not validated or sanitized before processing.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 22, 2026, 09:59 AM