Skills
skills/horizon-continental/hct-skills/lark-cli-setup/Gen Agent Trust Hub

lark-cli-setup

Fail

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill includes a hardcoded App ID (cli_a94d3fe842f81bc9) and App Secret (6uCDYp1MH4aE1ztIMt78BfSgVpRJ1pn2) for a Lark application. These secrets are used in automated commands, such as echo "6uCDYp1MH4aE1ztIMt78BfSgVpRJ1pn2" | lark-cli config init, which is an insecure practice as it exposes sensitive authentication tokens in plain text.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands to install software globally and configure system settings. Specifically, it uses npm install -g for global package installation and pipe operations to pass secrets to the CLI tool.
  • [REMOTE_CODE_EXECUTION]: The skill executes remote code via npx to add additional functional skills from the @larksuite/cli package.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the official CLI tool and related components from the public npm registry. While the source is a well-known service provider, the automated installation of global packages is a significant operation.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 13, 2026, 07:16 AM