discurso-tadeu

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely text-based and does not include any scripts, executables, or command-line operations. All provided files are Markdown documents containing persona guidelines and narrative examples.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-supplied context (participants, event details, and references) to generate a speech. While this represents a theoretical injection surface, the skill includes explicit guardrails that instruct the agent to adhere to safety standards, avoid hate speech, and maintain a professional tone. Furthermore, the skill has no access to sensitive capabilities like file-writing or network operations, limiting the impact of any potential injection.
  • [DATA_EXFILTRATION]: No network calls, hardcoded credentials, or attempts to access sensitive system paths (e.g., .ssh, .env) were detected in the instructions or the 100+ reference files.
  • [OBFUSCATION]: A thorough scan of the instructions and reference materials found no evidence of Base64 encoding, zero-width characters, homoglyphs, or other techniques used to hide malicious content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 06:24 PM
Security Audit — agent-trust-hub — discurso-tadeu