devil-ai@0.14.1
Devil AI — AI-powered coding assistant. Install everything with: npm install -g devil-ai
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThis is a dangerous explicit-user-command agent bridge that modifies a related agent configuration and executes remote installers. The postinstall hook is inert, so the inspected source does not establish malicious automatic install behavior.
Decision evidence
public snapshot- The explicit start command modifies an existing OpenCode or DevilCode configuration file.
- A Telegram webhook forwards received messages to a local OpenCode agent, creating a remote control path to that agent.
- The explicit setup command executes remote installer scripts through a shell, including one with elevated Linux commands.
- The install lifecycle only prints a message and does not execute the CLI.
- No self-dependency, hidden payload, credential harvesting, or automatic network activity was found.
- The risky behavior requires an explicit CLI command and a user-provided Telegram bot token.
Source & flagged code
11 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to devil-ai@0.14.1.
See all versions of devil-ai for other recorded verdicts.
- Review the evidence and your use of devil-ai@0.14.1 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.