dojo-chris-voss

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists entirely of high-quality instructional Markdown files detailing negotiation frameworks and persona guidelines. No malicious intent, obfuscation, or unauthorized access patterns were detected.- [NO_CODE]: The skill does not distribute any scripts, binaries, or dependency manifests (e.g., package.json or requirements.txt). All logic is conveyed through natural language instructions to the AI agent.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided negotiation scenarios, which represents a potential ingestion surface for untrusted data.
  • Ingestion points: User questions about specific situations or requests for document reviews (SKILL.md).
  • Boundary markers: Instructions explicitly require the agent to load specific topic files and use only the substance within them, effectively scoping the response.
  • Capability inventory: The skill possesses no advanced capabilities such as network requests, file system writes, or subprocess execution.
  • Sanitization: No explicit input filtering is defined, though the strict adherence to provided topic files acts as a functional constraint.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 03:31 PM
Security Audit — agent-trust-hub — dojo-chris-voss