@lightupai/polaris@0.0.140
Record and stream Claude Code sessions to Slack. Like Gong for AI coding — capture every prompt, response, and tool call. Multiplayer collaboration across AI coding agents.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe flagged credential flow uses the app’s configured Slack bot token with Slack’s own API for signup notifications, and the flagged shell operations belong to explicit CLI setup or performance-audit commands. The manifest declares no install lifecycle scripts, and no unrelated credential collection or concealed execution was found in the reviewed paths.
Decision evidence
public snapshot- The package describes recording Claude Code sessions and streaming them to Slack.
- Signup notifications use the configured SIGNUP_SLACK_BOT_TOKEN and send to Slack’s API.
- The flagged npm install runs during the CLI’s default setup command; the package has no install lifecycle scripts.
- The performance audit runs shell commands only when explicitly run, and its local-server build is conditional on the `local` argument.
- The Slack notification sends signup and plan details to a fixed internal channel, which is a privacy-relevant application behavior.
- The CLI setup installs an MCP server and writes Claude Code configuration when the user runs setup.
Source & flagged code
6 flaggedPackage source references child process execution.
scripts/perf-audit.tsView on unpkg · L14A single source file combines environment access, network access, and code or shell execution; review context before blocking.
scripts/perf-audit.tsView on unpkg · L33Findings
Affected versions and remediation
This report applies to @lightupai/polaris@0.0.140.
See all versions of @lightupai/polaris for other recorded verdicts.
- Review the evidence and your use of @lightupai/polaris@0.0.140 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.