instar@1.3.1343
Coherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe source supports an automatically triggered, first-party agent configuration and hook migration after updates, which warrants a warning under the agent integration policy. The inspected evidence does not establish credential theft, a backdoor, or another concrete attack, and the postinstall hook is a native dependency repair.
Decision evidence
public snapshot- The package exposes a postinstall hook that runs a native SQLite repair script.
- The server runs migrations automatically after detecting an installed version change.
- The migrator states that it overwrites generated hooks, and an update-check fallback can invoke it.
- Reviewed the cited agent-integration locations; they show setup, migration, or project-memory behavior.
- The postinstall hook is scoped to repairing the better-sqlite3 native binary.
- The GitSync excerpt builds a git allowed-signers file from active machine identities.
- No inspected evidence shows credential exfiltration, a backdoor, or a payload loaded from the flagged hidden path.
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@1.3.1343.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1343 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.