epf-bsp-init

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by design, as it takes user input and injects it into code templates.
  • Ingestion points: Command arguments ProcessorName, Вид, and Назначение enter the agent context via user input.
  • Boundary markers: There are no boundary markers or instructions to treat user-provided data as untrusted, increasing the risk of the agent executing malicious instructions embedded in arguments.
  • Capability inventory: The skill uses Glob and Edit tools to modify local files (ObjectModule.bsl).
  • Sanitization: No input validation, escaping, or filtering is performed on the user-supplied strings before they are written to the source code.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 11:10 AM
Security Audit — agent-trust-hub — epf-bsp-init