skills/lilmgenius/paperthin/dedash/Gen Agent Trust Hub

dedash

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a specialized text processing utility for grammatical cleanup. It identifies and replaces various types of dashes (such as U+2014, U+2015, and specific en-dash usage) with punctuation like commas, colons, or semicolons based on the grammatical role in the sentence.
  • [DATA_EXPOSURE]: No patterns for accessing sensitive system files, environment variables, or credentials were detected. The skill explicitly limits its operations to a user-defined scope (a passage, field, file, or tree) and instructs the agent not to widen this scope without permission.
  • [PROMPT_INJECTION]: The instructions are focused on linguistic analysis and voice preservation. No attempts to bypass safety filters, override system instructions, or extract system prompts were found.
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute external code. The frontmatter includes a configuration disable-model-invocation: true which restricts the agent's ability to trigger other model-based tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 09:12 AM
Security Audit — agent-trust-hub — dedash