registry / @drdeeks/character-kit / 1.7.0
@drdeeks/character-kit@1.7.0
Agent character enforcement. Host-neutral core plus OpenAI, Claude, and Hermes companions. Fail-closed. Not a security boundary.
AI Security Review
scanned 6d ago · by lpm-firewall-aiSource inspection found a gated postinstall that only reports how to configure the package, while agent hook configuration is handled by a separate installer. The inspected cleanup and runtime helper behavior did not establish malicious activity.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source
Trigger
Installing the package runs its postinstall hook; configuring Claude hooks requires running the installer command.
Impact
No unconsented install-time agent configuration change or other concrete attack effect was established.
Mechanism
The postinstall hook checks install context, detects harness markers, and prints instructions. The installer writes the kit hook entries to Claude settings.
AI rationale
Source inspection found a gated postinstall that only reports how to configure the package, while agent hook configuration is handled by a separate installer. The inspected cleanup and runtime helper behavior did not establish malicious activity.
Decision evidence
public snapshotAI called this Clean at 95.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package lifecycle hooks call postinstall and preuninstall scripts; postinstall is limited to global installs from a node_modules tree and prints setup guidance without configuring harnesses.
- Claude hook configuration is written by the separate installer command; uninstall cleanup removes only matching kit hook entries and is similarly limited to global package installs.
- The inspected runtime hook delegates to the ack CLI, the harness detector checks local marker paths, the knowledge indexer excludes agent-internal files, and the watchdog checks process liveness.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStrings
Source & flagged code
10 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node node/bin/postinstall.js
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg•scripts.postinstall = node node/bin/postinstall.js
Medium
Ambiguous Install Lifecycle Script
Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgplugins/claude/hooks/ack-stdio.jsView file
1import { spawn } from "node:child_process";
L2: import { existsSync } from "node:fs";
High
Child Process
Package source references child process execution.
plugins/claude/hooks/ack-stdio.jsView on unpkg · L1Findings
4 Critical5 High5 Medium4 Low
CriticalSame File Env Network Executionnode/bin/install.js
CriticalAi Agent Control Hijacknode/src/detect-harnesses.js
CriticalTrigger Reachable Dangerous Capabilitynode/bin/install.js
CriticalPrevious Version Dangerous Deltanode/bin/install.js
HighInstall Time Lifecycle Scriptspackage.json
HighChild Processplugins/claude/hooks/ack-stdio.js
HighShell
HighSemantic Analysis Limitednode/bin/ack.js#virtual:normalized:round1
HighKnown Malware Source Similaritydeploy/ack_watchdog.js
MediumAmbiguous Install Lifecycle Scriptpackage.json
MediumNetwork
MediumEnvironment Vars
Affected versions and remediation
This report applies to @drdeeks/character-kit@1.7.0.
See all versions of @drdeeks/character-kit for other recorded verdicts.
- Review the evidence and your use of @drdeeks/character-kit@1.7.0 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.