instar@1.3.1401
⚠ Under reviewCoherence infrastructure for self-evolving AI agents — on the Claude Code or Codex subscription you already have.
AI Security Review
scanned 3d ago · by lpm-firewall-aiInspected source resolved several high-risk scanner hints to ordinary Telegram API, Git signing, and credential-prompt guard behavior. The requested clean review requires more citations than the 256-reference limit permits, so the remaining uncertainty is review coverage rather than confirmed malicious behavior.
Decision evidence
public snapshot- The manifest's postinstall runs a native SQLite repair script.
- The inspected Telegram request uses the Telegram Bot API and redacts the bot token in its safe URL.
- Git sync builds its allowed-signers file from active machine identities and configures Git to use it.
- The inspected hook blocks free-text credential prompts unless they are clearly multi-choice decisions.
- The migrator installs that free-text guard as an Instar hook.
- The inspected init code installs built-in skills under the project’s .claude directory.
Source & flagged code
17 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@1.3.1401.
See all versions of instar for other recorded verdicts.
- Review the evidence and your use of instar@1.3.1401 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.