negotiation-playbook
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Overall Assessment] (SAFE): The skill is a pure content-based expert system for negotiation. It does not include any scripts, executable code, or external dependencies.
- [Prompt Injection] (SAFE): The instructions in
SKILL.mdare standard prompt engineering designed to ground the AI in specific reference files. Phrases like "Ignore generic approaches" are restricted to choosing local reference patterns over default model knowledge and do not constitute a malicious bypass of safety filters. - [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or network-enabled commands (like curl or fetch) are present in any of the files.
- [Indirect Prompt Injection] (LOW): While the skill is designed to process user-provided negotiation scenarios (an ingestion point for untrusted data), it lacks any high-risk capabilities such as file writing, command execution, or network requests, effectively neutralizing the risk of a successful injection attack.
Audit Metadata