@netnodeag/kraftwerk@0.16.0
Deterministic workflow-as-code framework: agents (persona + model + tools + harness) run in bounded phases on headless CLI harnesses (claude -p, codex exec, pi); code owns the control flow, envelopes + gates judge the results
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package contains a real unattended agent-permission capability, so it warrants a warning as a dangerous automation tool. It has no registry-install hook or concrete malicious payload behavior.
Decision evidence
public snapshot- Scheduled team routines automatically approve an available agent permission option.
- User-selected workflow script phases execute their configured Bash text.
- Sandbox runs can pass API credentials and, when requested, an SSH agent into a container.
- There is no preinstall, install, or postinstall hook; the only lifecycle script is prepublishOnly.
- The risky operations are part of an agent-workflow CLI and require a user command or configured routine.
- No source evidence shows secret collection, covert exfiltration, self-dependency, or foreign agent-config mutation.
Source & flagged code
5 flaggedPackage source references dynamic require/import behavior.
bin/kraftwerk.jsView on unpkg · L15Findings
Affected versions and remediation
This report applies to @netnodeag/kraftwerk@0.16.0.
See all versions of @netnodeag/kraftwerk for other recorded verdicts.
- Review the evidence and your use of @netnodeag/kraftwerk@0.16.0 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.