encode-repo-serena

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted codebase data to populate a knowledge base, which presents an indirect prompt injection surface.
  • Ingestion points: Source code and documentation are read in references/phases.md using mcp__plugin_serena_serena__read_file and mcp__plugin_serena_serena__search_for_pattern.
  • Boundary markers: There are no explicit instructions or delimiters defined to separate codebase content from agent instructions during the encoding process.
  • Capability inventory: The skill uses execute_forgetful_tool to create, update, and delete memories and entities based on the processed data.
  • Sanitization: Content extracted from the codebase is interpolated directly into knowledge base tools without validation or sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 03:46 PM
Security Audit — agent-trust-hub — encode-repo-serena