skills/marclelamy/skills/multi-debate/Gen Agent Trust Hub

multi-debate

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to ingest peer debate contributions and incorporate their logic into the agent's reasoning.
  • Ingestion points: The agent reads content from all markdown files matching the turn-{N-1}-*.md pattern within a shared directory provided by the user in SKILL.md.
  • Boundary markers: No delimiters or isolation instructions are present to distinguish between the untrusted content of the external files and the agent's instructions.
  • Capability inventory: The agent uses directory listing (via ls) and file read/write operations to execute the workflow.
  • Sanitization: Lacks any mechanism to validate, filter, or sanitize the input from the external files before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 12:55 PM
Security Audit — agent-trust-hub — multi-debate