psyclaw@0.31.3
A standalone, evidence-grounded social-science research agent with research contracts, evidence gates, and recoverable workflows
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe package discloses default-on telemetry and offers an opt-out, but error reports can still transmit local path details that may identify project locations. This is a privacy risk warranting a warning, not evidence of credential theft or malicious intent.
Decision evidence
public snapshot- CLI use enables default-on product telemetry, with a telemetry-off command and an interactive notice.
- The CLI initializes telemetry at startup and includes the current working directory when reporting errors.
- The telemetry allowlist includes cwd and failed_path fields.
- PostHog error events transmit cwd and failed_path values, each up to 240 characters.
- Path redaction replaces the home-directory prefix with ~, but does not remove the remaining path.
- The manifest has no install, preinstall, or postinstall hook; prepublishOnly runs branding checks and a build.
- The LLM telemetry code says it omits prompts, completions, and tool I/O.
- The notice describes product-use and error telemetry and provides a way to disable it.
Source & flagged code
6 flaggedPackage source references dynamic require/import behavior.
dist/apps/panel/observability.jsView on unpkg · L39Findings
Affected versions and remediation
This report applies to psyclaw@0.31.3.
See all versions of psyclaw for other recorded verdicts.
- Review the evidence and your use of psyclaw@0.31.3 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.