cheat-status

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute line counting operations via the 'wc -l' command on local files like 'rubric_notes.md'.
  • [PROMPT_INJECTION]: An indirect prompt injection surface is identified where malicious instructions could be embedded in the files the skill processes.
  • Ingestion points: The skill reads project-specific files including '.cheat-state.json', 'predictions/*.md', 'candidates.md', and 'rubric_notes.md'.
  • Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are present to protect the agent from content inside these files.
  • Capability inventory: The agent has access to 'Bash(*)', 'Read', 'Glob', and 'Grep', which could be leveraged if the agent follows instructions found in the ingested data.
  • Sanitization: There is no evidence of validation or sanitization of the content retrieved from the file system before it is incorporated into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 06:58 PM