github

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • REMOTE_CODE_EXECUTION (CRITICAL): The skill documentation explicitly directs users to run curl -sSL https://canifi.com/skills/github/install.sh | bash and curl -sSL https://canifi.com/install.sh | bash. Pining remote scripts from an untrusted source (canifi.com) directly to a shell is a major security vulnerability that enables complete system compromise.
  • EXTERNAL_DOWNLOADS (HIGH): The skill relies on external scripts and environment management tools hosted on canifi.com, which is not a recognized trusted repository or organization. This introduces a significant supply-chain risk.
  • COMMAND_EXECUTION (HIGH): The installation and setup process involves executing shell commands that modify the local filesystem and environment, such as cp -r skills/github ~/.canifi/skills/, using logic provided by the unvetted external scripts.
  • CREDENTIALS_UNSAFE (MEDIUM): The documentation encourages users to store sensitive GitHub passwords in canifi-env for automated login. In conjunction with the untrusted installation scripts, this creates a high risk of credential harvesting and unauthorized account access.
Recommendations
  • HIGH: Downloads and executes remote code from: https://canifi.com/skills/github/install.sh, https://canifi.com/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 17, 2026, 05:37 PM
Security Audit — agent-trust-hub — github