lambda-lang

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a domain-specific language for agent-to-agent communication, creating an indirect prompt injection surface for data parsed from external agents.
  • Ingestion points: Agent-to-agent messaging channels and coordination pipelines (SKILL.md).
  • Boundary markers: Syntax utilizes specific prefixes (e.g., !, ?, #) to delineate intent and scope.
  • Capability inventory: No scripts, subprocess calls, or network operations are defined within this skill.
  • Sanitization: The documentation explicitly warns to treat user strings as untrusted and provides guidance on escaping them.
  • [SAFE]: The skill is entirely instructional and does not include any functional code, package dependencies, or executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 06:57 AM