registry / md-redline / 0.13.0
md-redline@0.13.0
Inline review comments for markdown specs. Built-in MCP server hands feedback directly to your AI agent.
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe manifest maps the package commands to the CLI and runs a build during prepublish. The command-output-exfiltration hint is explained by local Claude CLI output being checked for an existing registration; no external recipient or secret transfer was established.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review
Trigger
Running the CLI or using its MCP installation subcommand.
Impact
No attacker-directed data transfer or harmful effect was established.
Mechanism
The inspected network requests target the loopback server; the Claude CLI output is read to check whether this package is already registered.
AI rationale
The manifest maps the package commands to the CLI and runs a build during prepublish. The command-output-exfiltration hint is explained by local Claude CLI output being checked for an existing registration; no external recipient or secret transfer was established.
Network endpoints1
127.0.0.1
Decision evidence
public snapshotAI called this Clean at 88.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest maps the command to the CLI and limits its prepublish hook to a build; inspected CLI requests use loopback, and the Claude CLI output is checked for an existing registration.
- The flagged client bundles begin with imports linking application entry points to diagram, parser, and color utilities.
- The inspected dependency manifests identify font packages, and their CSS references local font files.
- The server binds to the loopback address.
Evidence against
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetworkShell
HighEntropyStringsMinifiedPossibleObfuscationUrlStrings
Source & flagged code
12 flaggedbin/server-control.jsView file
1import { execSync } from 'child_process';
L2:
High
Child Process
Package source references child process execution.
bin/server-control.jsView on unpkg · L1bin/cli.jsView file
310stdio: 'ignore',
L311: shell: true,
L312: windowsHide: true,
High
4Cross-file remote execution chain: bin/cli.js spawns dist/server.js; helper contains network access plus dynamic code execution.
L4: * The published entry point is `bin/md-redline`, which exists only to import
L5: * this file. That name is what `package.json` maps both commands to, and it
L6: * cannot carry an extension; without one, `tsconfig.bin.json` could not include
...
L14:
L15: import { connect } from 'net';
L16: import { spawn } from 'child_process';
L17: import { readdirSync, statSync } from 'fs';
...
L41: const __filename = fileURLToPath(import.meta.url);
L42: const __dirname = dirname(__filename);
L43:
...
L127: try {
L128: const parsed = JSON.parse(raw);
Findings
1 Critical10 High4 Medium6 Low
CriticalTrojan Source Unicodedist/assets/chunk-K5T4RW27-B-Tk5uU5.js
HighChild Processbin/server-control.js
HighShellbin/cli.js
HighSame File Env Network Executionbin/cli.js#virtual:normalized:round1
HighCommand Output Exfiltrationbin/cli.js#virtual:normalized:round1
HighCross File Remote Execution Contextbin/cli.js
HighTrigger Reachable Command Output Exfiltrationbin/cli.js#virtual:normalized:round1
HighSemantic Analysis Limitedbin/cli.js#virtual:normalized:round1
HighKnown Malware Source Similaritydist/assets/chunk-OYMX7WX6-Czzw3fO9.js
HighKnown Malware Source Similaritydist/assets/diagram-G4DWMVQ6-BGW_rFMi.js
Affected versions and remediation
This report applies to md-redline@0.13.0.
See all versions of md-redline for other recorded verdicts.
- Review the evidence and your use of md-redline@0.13.0 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.