tk-engineering

Pass

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown-based engineering guidelines, configuration files, and licensing information. There is no executable code (scripts, binaries, or active logic) included in the skill package.
  • [DATA_EXFILTRATION]: The skill provides explicit instructions to prevent accidental data exfiltration and credential exposure. It mandates that agents exclude passwords, tokens, OTPs, cookies, and other sensitive inputs from logs, reports, and diagnostic traces (found in references/diagnosis.md and references/sdd-transport.md).
  • [PRIVILEGE_ESCALATION]: The instructions establish a principle of least privilege for delegated tasks, requiring that child processes or agents have the same or narrower access than the parent agent. It specifically forbids automatic privilege escalation or the bypass of security filters (found in references/delegation-permissions.md).
  • [INDIRECT_PROMPT_INJECTION]: Several reference files are dedicated to identifying and mitigating indirect prompt injection risks. The skill instructs the agent to audit for malicious instructions in tool outputs, search results, and code changes that might attempt to override safety guardrails (found in references/security-boundaries.md and references/project-rules.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 15, 2026, 01:26 AM