skills/958877748/skills/cocos-build/Gen Agent Trust Hub

cocos-build

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes curl.exe to send a request to http://localhost:7456/update-db, which is the local API endpoint for Cocos Creator to refresh its asset database. This is a routine developer workflow operation.
  • [INDIRECT_PROMPT_INJECTION]: The skill monitors and acts upon files within the assets/ directory. Any tool that ingests files created or modified by users or external sources has a theoretical attack surface for indirect prompt injection, though here the risk is limited to the local development environment.
  • Ingestion points: Reads files from the assets/ directory (SKILL.md).
  • Boundary markers: None identified.
  • Capability inventory: Executes shell commands via curl.exe and interacts with a local web service.
  • Sanitization: None identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:43 AM
Security Audit — agent-trust-hub — cocos-build