fuck-slop

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use standard local shell utilities like grep, tr, sed, and awk for text analysis. These commands are used for their intended purpose of pattern matching and statistical measurement of text files.
  • [PROMPT_INJECTION]: As the skill is designed to process user-supplied text for rewriting, it contains an indirect prompt injection surface.
  • Ingestion points: User-provided text or documents for analysis.
  • Boundary markers: None present; input text is not isolated with specific delimiters.
  • Capability inventory: The agent can execute local shell commands and rewrite text content.
  • Sanitization: The skill does not describe specific methods for sanitizing or filtering instructions that might be embedded in the input text.
  • [SAFE]: No evidence of remote code execution, credential theft, network exfiltration, or obfuscation was found in the skill's instructions or supporting reference files. The skill operates within expected bounds for a text-editing tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 08:11 AM
Security Audit — agent-trust-hub — fuck-slop