skills/lelborn/skills/keep-it-human/Gen Agent Trust Hub

keep-it-human

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill contains strong behavioral directives such as 'The one rule that overrides the rest', but these are focused on maintaining factual accuracy and stylistic consistency rather than bypassing safety guardrails. The instruction to trigger even without specific keywords is a functional heuristic for the agent's behavior, not a malicious override.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external user-provided text, which creates an ingestion surface. However, the skill lacks access to sensitive capabilities like the file system, network, or shell execution, which limits the risk to text manipulation. The instructions explicitly mandate preserving all facts, caveats, and citations, which serves as a safeguard against accidental or intentional data distortion.
  • [SAFE]: No malicious code, obfuscation, persistence, or data exfiltration patterns were detected. The skill operates entirely within the scope of natural language processing without external dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 01:56 PM