instar@1.3.1379
⚠ 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 source did not establish malicious behavior, but the required clean citation set contains 423 paths and exceeds the 256-citation limit. Manual review is needed to resolve the remaining source coverage; citation capacity alone is not evidence of an attack.
Decision evidence
public snapshot- The manifest defines a postinstall script for a local SQLite fix, while its prepare and prepublishOnly hooks do not establish a malicious install chain.
- The migration CLI explicitly runs PostUpdateMigrator, which writes an Instar-named free-text guard hook.
- The inspected guard blocks credential-style free-text questions; the inspected excerpts did not establish an attack.
- The required clean review covers 423 distinct paths, exceeding the 256 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@1.3.1379.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1379 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.