instar@1.3.1402
⚠ Under reviewCoherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe inspected source does not confirm the scanner's credential redirect or persistence claims, and several observed behaviors align with the package's CLI and service functions. The full runtime review remains incomplete, so the source does not support a confident clean or malicious verdict.
Decision evidence
public snapshot- The manifest's postinstall runs a native-module repair script; prepare runs Husky.
- The inspected init and migrate paths place skills and hooks under project .claude directories as part of CLI setup or migration.
- The credential provider stores credentials in the operating system keychain, and the Telegram adapter sends Telegram API requests to api.telegram.org.
- The suspected GitSync persistence code writes a Git signing allowlist from active machine identities.
- The inspected source does not establish the scanner's claimed credential redirect or unauthorized persistence behavior.
- The available source excerpts do not resolve all flagged runtime behavior across this very large package.
Source & flagged code
17 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
dist/threadline/PipeSessionSpawner.jsView on unpkg · L15Package source references shell execution.
Findings
Affected versions and remediation
This report applies to instar@1.3.1402.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1402 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.