event-accessibility-inclusion

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown instructions and reference documentation. No executable scripts, shell commands, or network-enabled tools are present in the skill files.
  • [DATA_EXFILTRATION]: No network access or sensitive file reading patterns were detected. The skill provides explicit best-practice guidance for the secure handling, restricted access, and timely deletion of attendee disability disclosures.
  • [PROMPT_INJECTION]: No evidence of prompt injection attempts, safety filter bypasses, or instructions to ignore safety guidelines was found in the instructions or metadata.
  • [OBFUSCATION]: No base64 encoding, hex escapes, zero-width characters, homoglyphs, or other obfuscation techniques were identified in the text or references.
  • [INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided event details, it lacks any dangerous capabilities (such as shell execution, file-writing, or network operations) that could be exploited via malicious input from third-party event descriptions.
  • [DYNAMIC_CONTEXT_INJECTION]: No use of shell command execution at load time (exclamation mark plus backtick pattern) was detected in the SKILL.md or reference files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 09:13 PM
Security Audit — agent-trust-hub — event-accessibility-inclusion