warden-ai
Verified context for AI coding agents. A local context layer that intelligently compresses, filters, indexes, and verifies AI coding-agent tool output. 32 MCP tools across 10 layers: code intelligence (tree-sitter, 30+ languages), context selection, tool
This report shows the evidence for warden-ai@0.7.1. 23 of 23 known versions are blocked or flagged. See all versions.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has a concrete automatic lifecycle chain that silently changes foreign AI-agent configuration and tool controls. This meets the install-control-surface blocking policy.
Decision evidence
public snapshot- The manifest automatically runs a postinstall script.
- The postinstall script silently invokes init in the consumer's current directory.
- Init registers a server in global agent configurations and enables automatic tool approval.
- Init writes PreToolUse hooks into consumer project agent settings.
- No credential collection or outbound network send was identified in the inspected execution path.
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 unpkgFindings
Affected versions and remediation
This report applies to warden-ai@0.7.1.
The versions list below shows the verdict for each recorded version.
- Avoid installing warden-ai@0.7.1. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.