Skills
skills/algorand-devrel/algorand-agent-skills/create-python-x402-facilitator-bazaar/Gen Agent Trust Hub

create-python-x402-facilitator-bazaar

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • EXTERNAL_DOWNLOADS (MEDIUM): The skill instructs the user to install the x402-avm package via pip from an untrusted source. The GitHub organization GoPlausible is not on the trusted list.
  • PROMPT_INJECTION (MEDIUM): The skill contains an indirect prompt injection surface (Category 8).
  • Ingestion points: The extract_discovery_info function in SKILL.md (Step 5) ingests payment_payload and payment_requirements from untrusted external resource servers.
  • Boundary markers: The code uses a validate=True flag, but does not implement explicit delimiters to prevent malicious instructions from being interpreted as data during cataloging.
  • Capability inventory: The facilitator side builds API catalogs and prints descriptions. While no immediate RCE is shown, the aggregated data can influence downstream agent decisions or user behavior.
  • Sanitization: Uses jsonschema for structural validation, which prevents schema confusion but does not filter for natural language instructions embedded in fields like 'description'.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 08:51 AM