@antoneeo/course-creator@0.1.0-beta.4
Design sourced, progressive courses for a defined learner with Agentic SDLC governance and diagnostic learner simulations.
AI security review
scanned · by lpm-firewall-aiThe automatic global Claude session integration creates a concrete agent extension risk, so a warning is appropriate. The inspected hook is package-owned and limited to reading project documentation; the source does not establish a publish-blocking attack.
Decision evidence
public snapshot- The manifest runs scripts/postinstall.js during npm installation.
- Postinstall installs the skill for detected assistant clients and, after a new Claude skill install, attempts to add a global orientation hook.
- The hook adds a command to Claude’s global SessionStart settings. The orientation command reads a bounded set of project documents and returns without subprocess execution.
- The explicit project initializer writes agent guidance only when the destination file is absent.
- The install path copies this package’s own skill, and the hook points to that installed skill’s orientation command.
- The orientation command reads project documentation into session output; inspected source did not show credential collection or network transfer.
Findings
1 critical · 3 high · 5 medium · 3 lowWhere they are, most severe file first. Code opens a file's flagged lines, with their evidence, below.
scripts/lib.js1 criticalpackage.json1 high · 2 mediumscripts/init.js1 highscripts/lib.js#virtual:normalized:round11 highskills/course-creator/scripts/sdlc_check.py1 mediumShow all 12 findings
scripts/lib.js1 findingpackage.json3 findingsscripts/init.js1 findingscripts/lib.js#virtual:normalized:round11 findingskills/course-creator/scripts/sdlc_check.py1 findingSource & flagged code
7 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgShared scanner record · 2 findings · 1 line
The manifest runs scripts/postinstall.js during npm installation.
package.jsonView on unpkg · L14Code excerpt · 3 lines
Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.
scripts/lib.jsView on unpkg · L75Code excerpt · 12 lines
Manifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.
scripts/init.jsView on unpkgScanner record · 4 lines
Manifest-trigger-reachable source links an external AI-agent control path to a behavior-bearing write operation.
Package ships non-JavaScript build or shell helper files.
skills/course-creator/scripts/sdlc_check.pyView on unpkgScanner record · 4 lines
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
scripts/lib.js#virtual:normalized:round1View on unpkgScanner record · 1 line
Affected versions and remediation
This report applies to @antoneeo/course-creator@0.1.0-beta.4.
See all versions of @antoneeo/course-creator for other recorded verdicts.
- Review the evidence and your use of @antoneeo/course-creator@0.1.0-beta.4 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.