instar@1.3.1370
⚠ Under reviewCoherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe inspected scanner-highlighted excerpts did not establish malicious behavior, and some align with ordinary package functions. The required clean-source coverage cannot be completed within the citation limit, so the verdict remains unresolved.
Decision evidence
public snapshot- package.json runs a native SQLite repair script during postinstall; the inspected script describes checking or repairing that dependency.
- The inspected free-text guard describes blocking credential requests through a specific agent input tool.
- GitSync writes machine signing public keys to Git's allowed-signers file.
- The inspected migrator excerpt dynamically loads filesystem, path, and child-process modules for a project hook.
- The inspected excerpts do not establish credential theft, unauthorized persistence, or another concrete attack.
- The required clean review spans more source paths than can be covered within the citation limit.
Source & flagged code
18 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@1.3.1370.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1370 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.