Loading npm security reports…

@houseofwolvesllc/claude-scrum-skill@2.8.0: AI-agent | LPM Firewall

registry / @houseofwolvesllc/claude-scrum-skill / 2.8.0

@houseofwolvesllc/claude-scrum-skill@2.8.0

Claude Code skills for scrum project management — PRD to production release pipeline with project scaffolding, sprint planning, status tracking, sprint releases, full-project emulation testing, autonomous orchestration, and project cleanup.

AI Security Review

scanned 4h ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. The npm postinstall hook modifies Claude Code skill directories without a separate user command. A global install targets the shared Claude skills directory and can replace matching skill files.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

Installing the package with npm, especially npm install -g.

Impact

It can add or overwrite Claude Code skills in the global user profile or consuming project, changing the agent instructions and executable workflow payload available to Claude Code.

Mechanism

Postinstall copies bundled skills into Claude Code control directories and imports copied workflow modules.

Policy narrative

npm invokes bin/install.js automatically after installation. The script selects either ~/.claude/skills for global installs or the consuming project's .claude/skills directory, then recursively copies its bundled skills, workflows, and guidance there. Matching destination files are overwritten. It subsequently dynamically imports copied workflow modules, so package-controlled code is executed during the lifecycle hook. This changes a broad AI-agent control surface without a separate setup command.

Rationale

This is an automatic postinstall mutation of Claude Code's global or project skill control surface, including overwriting matching skills and executing copied modules. The lack of network or credential theft does not neutralize the install-hook control-surface hijack.

Evidence

package.jsonbin/install.js~/.claude/skills/<project>/.claude/skills/<project>/.gitignore

Decision evidence

Source & flagged code

7 flagged · loading source

package.jsonView file

•scripts.postinstall = node bin/install.js

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

9"scripts": { "postinstall": "node bin/install.js", "test": "node --test" },

High

Suspicious Lifecycle Evidence

The package automatically runs its installer during npm postinstall.

package.jsonView on unpkg · L9

•scripts.postinstall = node bin/install.js

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

README.mdView file

•- **`--skip-on-pause`** (default off) — when a spec's orchestration pauses on a safety gate, mark it skipped and advance to the next spec instead of pausing the queue. Use with caution; safety gates exist for a reason.

High

Ai Reviewer Manipulation

Package text addresses the security reviewer or scanner and tries to influence the review outcome.

README.mdView on unpkg

bin/install.jsView file

27const skillsDir = resolveSkillsDir(); L28: const location = isGlobalInstall() ? 'global (~/.claude/skills/)' : `project (${skillsDir})`; L29: console.log(`\n📋 Installing claude-scrum-skill (${location})...\n`); L30: L31: fs.mkdirSync(skillsDir, { recursive: true }); L32: ... L55: if (isGlobalInstall()) { L56: return path.join(HOME, '.claude', 'skills'); L57: } L58: return path.join(resolveProjectRoot(PACKAGE_ROOT), '.claude', 'skills'); L59: } ... L184: const gitignorePath = path.join(projectRoot, '.gitignore');

Critical

Ai Agent Control Hijack

Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.

bin/install.jsView on unpkg · L27

2L3: const fs = require('fs'); L4: const path = require('path');

Medium

Dynamic Require

Package source references dynamic require/import behavior.

bin/install.jsView on unpkg · L2

lib/workflows/_shared/resolve_worktree_concurrency.mjs#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4

High

Semantic Analysis Limited

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

lib/workflows/_shared/resolve_worktree_concurrency.mjs#virtual:normalized:round1View on unpkg

Source & flagged code

7 flagged

package.jsonView file

Lines 1-30json

1{

2 "name": "@houseofwolvesllc/claude-scrum-skill",

3 "version": "2.8.0",

4 "description": "Claude Code skills for scrum project management — PRD to production release pipeline with project scaffolding, sprint planning, status tracking, sprint releases, full-project emulation testing, autonomous orchestration, and project cleanup.",

5 "bin": {},

6 "publishConfig": {

7 "access": "public"

8 },

9 "scripts": {

10 "postinstall": "node bin/install.js",

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg · L10

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg · L10

11 "test": "node --test"

12 },

High

Suspicious Lifecycle Evidence

The package automatically runs its installer during npm postinstall.

package.jsonView on unpkg · L9

13 "keywords": [

14 "claude",

15 "claude-code",

16 "skills",

17 "scrum",

18 "sprint",

19 "project-management",

20 "agile",

21 "github-projects"

22 ],

23 "author": "House of Wolves LLC",

24 "license": "Apache-2.0",

25 "repository": {

26 "type": "git",

27 "url": "https://github.com/houseofwolvesllc/claudescrumskill.git"

28 },

29 "homepage": "https://github.com/houseofwolvesllc/claudescrumskill#readme",

30 "bugs": {

README.mdView file

•- **`--skip-on-pause`** (default off) — when a spec's orchestration pauses on a safety gate, mark it skipped and advance to the next spec instead of pausing the queue. Use with caution; safety gates exist for a reason.

High

Ai Reviewer Manipulation

Package text addresses the security reviewer or scanner and tries to influence the review outcome.

README.mdView on unpkg

bin/install.jsView file

Lines 1-47javascript

1#!/usr/bin/env node

2

3const fs = require('fs');

4const path = require('path');

Medium

Dynamic Require

Package source references dynamic require/import behavior.

lib/workflows/_shared/resolve_worktree_concurrency.mjs#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4

High

Semantic Analysis Limited

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

lib/workflows/_shared/resolve_worktree_concurrency.mjs#virtual:normalized:round1View on unpkg