registry / AI-agent capability abuse / fraim / 2.0.315

fraim@2.0.315

FRAIM core CLI and MCP package.

AI Security Review

scanned 3d ago · by lpm-firewall-ai

The explicit setup command silently enables user-level Claude Code token telemetry across projects, creating a privacy-sensitive configuration change. The receiver is local and the inspected parser handles usage metrics, so the evidence supports a warning rather than a block.

Static reason

One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source

Trigger

The user runs init-project while Claude Code is detected, or invokes add-ide for Claude Code.

Impact

Claude Code usage metrics become available to the local FRAIM receiver across projects using that user-level configuration.

Mechanism

The setup helper adds telemetry environment settings to ~/.claude/settings.json, directing metrics to FRAIM's local receiver.

AI rationale

Evidence

dist/src/cli/commands/init-project.jsdist/src/cli/commands/add-ide.jsdist/src/cli/setup/claude-code-telemetry.jsdist/src/local-mcp-server/otlp-metrics-receiver.js~/.claude/settings.json

Network endpoints2

localhost:4318127.0.0.1:4318

Decision evidence

public snapshot

AI called this Suspicious at 87.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

The init-project command enables Claude Code token telemetry when Claude Code is detected; add-ide also calls the setup helper for Claude Code.
The helper adds telemetry settings to the user-level Claude settings file and writes the updated settings.
The telemetry settings enable Claude Code telemetry and route metrics to localhost:4318.
The local receiver parses token, cache, cost, session, and model metrics and listens only on 127.0.0.1.
The other inspected agent setup locations refer to FRAIM project or skill artifacts.

Evidence against

The inspected telemetry receiver is loopback-bound, and its parser extracts usage metrics rather than prompt or response content.

Behavioral surface

Source

ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork

Supply chain

HighEntropyStringsUrlStrings

Manifest

Source & flagged code

17 flagged

dist/src/core/utils/git-utils.jsView file

Lines 1-29javascript

1"use strict";

2Object.defineProperty(exports, "__esModule", { value: true });

3exports.isExplicitLocalRepoPath = isExplicitLocalRepoPath;

4exports.getPort = getPort;

5exports.determineDatabaseName = determineDatabaseName;

6exports.getCurrentGitBranch = getCurrentGitBranch;

7exports.determineSchema = determineSchema;

8exports.getDefaultBranch = getDefaultBranch;

9exports.sanitizeRepoIdentifier = sanitizeRepoIdentifier;

10const child_process_1 = require("child_process");

11function extractLocalFolderLabel(rawPath) {

High

Child Process

Package source references child process execution.

dist/src/core/utils/git-utils.jsView on unpkg · L9

12 const normalized = rawPath

13 .trim()

14 .replace(/^file:\/\//i, '')

15 .replace(/\\/g, '/')

16 .replace(/\/+$/, '');

17 if (!normalized)

18 return undefined;

19 const parts = normalized.split('/').filter(Boolean);

20 const candidate = parts[parts.length - 1];

21 return candidate || undefined;

22}

23function isExplicitLocalRepoPath(repoUrl) {

24 if (!repoUrl || typeof repoUrl !== 'string') {

25 return false;

26 }

27 const trimmed = repoUrl.trim();

28 return (trimmed.startsWith('/') ||

29 trimmed.startsWith('\\') ||

bin/fraim.jsView file

Lines 1-14javascript

1#!/usr/bin/env node

3try {

4 require('../dist/src/cli/fraim.js');

5} catch (error) {

Medium

Dynamic Require

Package source references dynamic require/import behavior.

bin/fraim.jsView on unpkg · L3

6 if (error && error.code

dist/src/first-run/session-service.jsView file

Lines 26-66javascript

26 };

27 return function (mod) {

28 if (mod && mod.__esModule) return mod;

29 var result = {};

30 if

dist/src/cli/commands/override.jsView file

Lines 1-30javascript

1"use strict";

2var __importDefault = (this && this.__importDefault) || function (mod) {

3 return (mod && mod.__esModule) ? mod :

dist/src/cli/setup/user-level-sync.jsView file

Lines 113-153javascript

113 pkg = JSON.parse(fs_1.default.readFileSync(pkgPath, 'utf8'));

114 }

115 catch {

116 // Corrupt package.json — overwrite below rather than crash.

117 pkg = {};

dist/src/cli/utils/managed-agent-install.js#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 7

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

dist/src/cli/utils/managed-agent-install.js#virtual:normalized:round1View on unpkg

index.jsView file

•matchType = normalized_sha256 matchedPackage = fraim@2.0.183 matchedPath = index.js matchedIdentity = npm:ZnJhaW0:2.0.183 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High