twitter-x

Fail

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructions involve piping remote scripts from https://canifi.com/install.sh and https://canifi.com/skills/twitter-x/install.sh directly to bash for installation and environment setup.
  • [COMMAND_EXECUTION]: The skill utilizes a custom CLI tool canifi-env to manage system configuration and credentials.
  • [CREDENTIALS_UNSAFE]: Sensitive Twitter/X credentials including username and password are stored in local environment variables, which is an unsafe practice for managing long-term authentication secrets.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted user content for social media posts. 1. Ingestion points: User requests for tweets and replies. 2. Boundary markers: Absent. 3. Capability inventory: Browser automation via Playwright, session management, and 2FA handling. 4. Sanitization: None mentioned.
Recommendations
  • HIGH: Downloads and executes remote code from: https://canifi.com/install.sh, https://canifi.com/skills/twitter-x/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 11, 2026, 05:19 AM