koishi-plugin-gl-bot@0.0.24
GleamSlime Koishi Rebot Plugins
AI Security Review
scanned 15d ago · by lpm-firewall-aiThe active message handler intentionally returns all environment variable names and values to the chat, creating a concrete secret disclosure path. This supports a malicious credential-exfiltration verdict.
Static reason
One or more suspicious static signals were detected.
Trigger
A message matches the fixed environment-variable phrase.
Impact
Chat participants may receive credentials and other secrets configured in the bot process.
Mechanism
The handler enumerates process.env, formats each name and value, and sends the result as a chat message.
Attack narrative
When the plugin is running, a matching message causes the handler to send the process environment into the current chat session. This can disclose credentials and other secrets to chat participants; no external web endpoint is needed for the disclosure.
AI rationale
The active message handler intentionally returns all environment variable names and values to the chat, creating a concrete secret disclosure path. This supports a malicious credential-exfiltration verdict.
Evidence
lib/gl/index.js
Decision evidence
public snapshotAI called this Malicious at 98.0% confidence as Malware with low false-positive risk.
Evidence for block
- On each message, the handler sends every environment variable to the chat when the message text matches a fixed phrase.
- The plugin registers that message handler during initialization, making the disclosure active at runtime.
- The manifest has no install lifecycle script; the exposure is in runtime message handling.
Evidence against
- The package manifest lists only formatting, lint, and build scripts, with no install lifecycle hook.
Behavioral surface
CryptoEnvironmentVarsEvalFilesystemNetworkWebSocket
HighEntropyStringsUrlStrings
Source & flagged code
4 flagged · loading sourcelib/utils/cat_enc.jsView file
7917patternName = generic_password
severity = medium
line = 7917
Medium
package.jsonView file
10"scripts": {
"format": "prettier --write .",
"lint": "eslint .",
"fl": "prettier --write . && eslint . --fix",
"build:console": "koishi-console build",
"build": "tsc -p tsconfig.test.json
High
Suspicious Lifecycle Evidence
The manifest has no install lifecycle script; the exposure is in runtime message handling.
package.jsonView on unpkg · L10lib/gl/index.jsView file
64globalCommand() {
this.ctx.on('message', session => {
if (constants_1.IS_DEV) {
// console.log(JSON.stringify(session, null, 2));
}
switch (session.content) {
case '环境变量':
session.send(`当前环境变量：${Object.entries(process.env)
Findings
3 High3 Medium5 Low
HighSuspicious Lifecycle Evidencepackage.json
HighAi Review Evidencelib/gl/index.js
HighAi Review Evidencelib/gl/index.js
MediumSecret Patternlib/utils/cat_enc.js
MediumNetwork
MediumEnvironment Vars
LowScripts Present
LowEval
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to koishi-plugin-gl-bot@0.0.24.
See all versions of koishi-plugin-gl-bot for other recorded verdicts.
- Avoid installing koishi-plugin-gl-bot@0.0.24. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.