shunk031-writing-telegraph

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The SKILL.md file contains an instruction for the agent to state a specific phrase ('πŸ“‘ 私は shunk031-writing-telegraph γ‚’θͺ­γΏγΎγ—γŸγ€‚') acknowledging that it has read the content. This is a functional directive rather than a malicious safety bypass.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external text input for rewriting. It does not implement delimiters or sanitization, creating a surface for indirect instructions. However, since the skill has no tool access (network, shell, or file system), this surface presents no significant security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 10:05 AM
Security Audit β€” agent-trust-hub β€” shunk031-writing-telegraph