reduce-similarities

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the similarity-rs command-line tool via Bash to perform AST-based code analysis. This tool is restricted by the skill's configuration to a specific allowed command set.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes Rust source files (**/*.rs), which presents a potential surface for indirect prompt injection if the files contain adversarial content designed to influence the agent's refactoring logic.\n
  • Ingestion points: External Rust source files ingested during the similarity analysis process.\n
  • Boundary markers: No explicit delimiters or boundary markers are defined to isolate untrusted code content from the agent's instruction context.\n
  • Capability inventory: The skill has access to file system read tools (Read, Grep, Glob) and the ability to execute the similarity-rs analysis tool.\n
  • Sanitization: There is no evidence of content sanitization or instruction-filtering for the code being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 07:21 AM
Security Audit — agent-trust-hub — reduce-similarities