skills/terrylica/cc-skills/mark-read/Gen Agent Trust Hub

mark-read

Warn

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a Python script via 'uv run'. It constructs the command by placing the user-provided 'chat' parameter directly into a shell string within a heredoc block.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through command injection. 1. Ingestion points: The user-supplied 'chat' parameter in SKILL.md. 2. Boundary markers: No delimiters or quotes are used to isolate the user input from the shell command. 3. Capability inventory: Access to the Bash tool allows for execution of arbitrary system commands. 4. Sanitization: The skill lacks any instructions for input validation or escaping to prevent the use of shell metacharacters like semicolons or backticks.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 4, 2026, 09:52 AM