negotiating-offers
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No patterns of instruction override, jailbreaking, or system prompt extraction were found. The instructional language is focused solely on the negotiation task.
- [Data Exposure & Exfiltration] (SAFE): There are no network operations or sensitive file path accesses. The skill specifically includes instructions in
SKILL.mdandreferences/INTAKE.mdto avoid requesting secrets and to allow the user to redact sensitive details. - [Unverifiable Dependencies & Remote Code Execution] (SAFE): No external packages or remote scripts are referenced or executed. The skill consists entirely of Markdown guidance.
- [Obfuscation] (SAFE): No encoded strings, zero-width characters, or homoglyphs were detected in any of the analyzed files.
- [Indirect Prompt Injection] (SAFE): Although the skill processes user-provided data (offer details), it lacks any executable capabilities (no shell access, no network, no file writes) that could be exploited via malicious data input.
Audit Metadata