psyclaw@0.30.18
A standalone, evidence-grounded social-science research agent with research contracts, evidence gates, and recoverable workflows
AI Security Review
scanned 19d ago · by lpm-firewall-aiThe manifest has no automatic install hook; its prepublishOnly script runs build checks. Inspection of the required flagged paths found test fixtures, provider-name reads, and explicit user-controlled local installation behavior, with no concrete attack surface established.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
The CLI launcher runs at invocation; plugin or skill installation requires an explicit manager action, and TUI installation requires confirmation.
Impact
The inspected behavior does not establish credential disclosure, unrequested account actions, or lifecycle mutation of foreign agent settings.
Mechanism
The launcher configures PsyClaw-specific directories. The panel reads authentication files to identify providers, while installation code writes locally selected plugins or managed skills.
AI rationale
The manifest has no automatic install hook; its prepublishOnly script runs build checks. Inspection of the required flagged paths found test fixtures, provider-name reads, and explicit user-controlled local installation behavior, with no concrete attack surface established.
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- package.json maps the CLI to scripts/psyclaw.mjs and its prepublishOnly hook runs branding checks and a build.
- The launcher sets PsyClaw-specific state directories before importing the CLI.
- The panel reads authentication files to identify configured providers; the inspected code records provider names rather than forwarding credentials.
- The extension installs a recommended plugin locally after an install action is selected.
- The TUI runs an installation only after the user confirms with Y.
- The flagged Python file contains pytest fixtures and YAML helpers, and recommended skills are installed into the package-managed project imports directory.
Evidence against
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsTelemetryUrlStrings
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/skills/recommended.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@0.30.18.
See all versions of psyclaw for other recorded verdicts.
- Review the evidence and your use of psyclaw@0.30.18 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.