@xfive/chisel-ai-toolkit@1.1.0
Chisel AI toolkit — skills, rules and reference docs for AI coding agents working on Chisel WordPress themes
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe automatic postinstall installs first-party agent configuration and a tool hook into the detected project without a separate consent prompt, so a warning is appropriate. The inspected behavior is narrowly scoped and does not establish a concrete attack.
Decision evidence
public snapshot- The package postinstall hook runs install.js automatically.
- The installer defaults to configuring Claude and general agent instructions, then calls their emitters unless dry-run is selected.
- The Claude emitter recursively replaces package-managed skill and reference directories under the project’s .claude directory.
- The installer adds a PreToolUse hook to the project’s Claude settings; the hook blocks edits inside the project’s core directory.
- The hook checks a requested file path and project directory, with no network request or credential transfer in its inspected behavior.
- Uninstall removes managed content and its own settings hook while preserving other settings entries.
- The installed hook has a narrow effect: it blocks edits inside core and allows other tool calls.
- The inspected setup code does not send data to a network endpoint or collect credentials.
Source & flagged code
6 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 unpkgThe package postinstall hook runs install.js automatically.
package.jsonView on unpkg · L24Findings
Affected versions and remediation
This report applies to @xfive/chisel-ai-toolkit@1.1.0.
See all versions of @xfive/chisel-ai-toolkit for other recorded verdicts.
- Review the evidence and your use of @xfive/chisel-ai-toolkit@1.1.0 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.