negotium@0.3.11
Install the Negotium multi-agent runtime and CLI with one package
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package contains concrete high-risk control of a foreign AI-agent session and permission model, so it warrants a warning. The inspected lifecycle downloader verifies hashes and no confirmed credential-redirection/exfiltration chain was found, so source facts do not justify a publish block.
Decision evidence
public snapshot- Runtime fabricates Claude session records under ~/.claude and replays bundled agent attachments.
- Fabricated Claude records mark user turns as bypassPermissions.
- Claude provider explicitly enables dangerously skipping permissions.
- Postinstall downloads and installs executable binaries.
- No fixed credential gateway or command-output exfiltration was confirmed.
- Postinstall downloads are SHA-256 verified and stored under the package-owned .negotium state directory.
- Snippet publishing is disabled unless an operator supplies its endpoint via environment.
Source & flagged code
10 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 unpkgPackage source references child process execution.
dist/chunk-0ynjwr50.jsView on unpkg · L23Findings
Affected versions and remediation
This report applies to negotium@0.3.11.
See all versions of negotium for other recorded verdicts.
- Review the evidence and your use of negotium@0.3.11 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.