instar@1.3.1422
⚠ Under reviewCoherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 6h ago · by lpm-firewall-aiInspected source shows a native dependency repair hook, gated telemetry, and setup-time agent configuration, with no confirmed attack in those paths. Full clean certification requires more source citations than the output limit allows, so the package needs manual review.
Decision evidence
public snapshot- package.json runs a postinstall script that checks and repairs the better-sqlite3 native binary.
- The repair script downloads the matching better-sqlite3 prebuild or rebuilds that dependency from source.
- Telemetry starts only when enabled and sends an installation identifier with aggregate usage metrics.
- The inspected setup code writes a Playwright MCP entry to Codex configuration when the setup flow runs.
- The inspected install hook and repair code show no collection or transfer of user credentials.
- The inspected telemetry startup has an enabled check; no install-time telemetry send was identified.
Source & flagged code
29 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@1.3.1422.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1422 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.