psyclaw@0.31.2
A standalone, evidence-grounded social-science research agent with research contracts, evidence gates, and recoverable workflows
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe reviewed entry point isolates the bundled runtime, and the flagged shell command is user-facing installation guidance rather than automatic execution. Runtime telemetry is disclosed with an option to disable it; no malicious behavior was identified in the inspected source.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Running the CLI starts its bundled agent; telemetry initializes only when enabled, and plugin installation requires a user selection.
Impact
No malicious collection, credential transfer, unrequested account action, or install-time execution was found in the reviewed paths.
Mechanism
The launcher isolates the bundled agent configuration under PsyClaw’s own config directory. Telemetry is described as anonymous usage and error reporting with an off control. Catalog installation commands are guidance text.
AI rationale
The reviewed entry point isolates the bundled runtime, and the flagged shell command is user-facing installation guidance rather than automatic execution. Runtime telemetry is disclosed with an option to disable it; no malicious behavior was identified in the inspected source.
Evidence
~/.psyclaw/agent/psyclaw-settings.json~/.psyclaw/agent/sessions
Network endpoints3
cdn.jsdelivr.net/npm/@sentry/browser@10/+esmcdn.jsdelivr.net/npm/posthog-js@1/+esmus.posthog.com (default; configurable)
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package exposes a CLI at scripts/psyclaw.mjs; no preinstall, install, or postinstall hook is declared.
- The launcher sets the bundled agent’s config and session directories under PsyClaw’s own configuration area before importing the CLI.
- Runtime telemetry is disclosed as anonymous product usage and error reporting, and users can disable it.
- The recommended plugin installer runs only after a user selects an install action in the manager.
- The shell installation command in the catalog is presented as guidance for an external tool, not as code executed by package startup.
Evidence against
- The manifest includes a prepublishOnly build script, which is not an install hook for registry tarball installs.
- The panel observability code loads analytics SDKs only when a DSN or project key is configured.
- The package includes a Windows executable and vendored test fixtures, but the inspected flagged fixture is ordinary pytest configuration.
Behavioral surface
ChildProcessCryptoDynamicRequire
Source & flagged code
6 flaggeddist/apps/panel/observability.jsView file
39async function loadEsm(url) {
L40: return import(url);
L41: }
Medium
Dynamic Require
Package source references dynamic require/import behavior.
dist/apps/panel/observability.jsView on unpkg · L39vendor/windows/x64/fd.exeView file
•path = vendor/windows/x64/fd.exe
kind = native_binary
sizeBytes = 4104192
magicHex = [redacted]
Medium
vendor/ars/tools/release-discipline/scripts/sync-toolkit.shView file
Findings
3 High6 Medium6 Low
HighPayload In Excluded Dirvendor/ars/scripts/adapters/tests/conftest.py
HighSemantic Analysis Limiteddist/src/tui/app.js#virtual:normalized:round1
HighPrevious Version Dangerous Deltadist/src/chat.js
MediumDynamic Requiredist/apps/panel/observability.js
MediumNetwork
MediumEnvironment Vars
MediumShips Native Binaryvendor/windows/x64/fd.exe
MediumShips Build Helpervendor/ars/tools/release-discipline/scripts/sync-toolkit.sh
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
Low
Affected versions and remediation
This report applies to psyclaw@0.31.2.
See all versions of psyclaw for other recorded verdicts.
- Review the evidence and your use of psyclaw@0.31.2 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.