creator-registry

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which introduces a vulnerability surface for indirect prompt injection. 1. Ingestion points: The skill is designed to process pasted records, audience exports, and creator proposal events as defined in SKILL.md. 2. Boundary markers: The procedure section explicitly instructs the agent to treat pasted records as untrusted evidence. 3. Capability inventory: The skill utilizes a local Python script to read from and append to a creator event stream (NDJSON) and regenerate Markdown views. 4. Sanitization: Instructions mandate the minimization of PII and require verified handle links before performing identity deduplication.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 07:29 PM
Security Audit — agent-trust-hub — creator-registry