instar@1.3.1416
Coherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 23h ago · by lpm-firewall-aiThe setup flow creates a first-party agent extension configuration and accepts project trust without a separate prompt. This warrants a warning, but the inspected source does not establish malicious execution or an install-hook attack.
Decision evidence
public snapshot- The setup flow registers Playwright MCP through `npx` using the latest release and marks project trust as accepted without a separate prompt.
- The manifest's postinstall runs a SQLite repair; the cited agent configuration changes are reached from the setup flow.
- The inspected credential paths use OS keychain commands, and the quota event handler queues a notification for an emitted enforcement event.
- The identified agent configuration change requires running the setup flow; the inspected manifest does not make that flow an npm install hook.
- The cited credential and quota code does not establish a separate attack.
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@1.3.1416.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1416 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.