Loading npm security reports…

@titan-design/active-work@0.11.0: AI-agent control-surface risk | LPM Firewall

registry / @titan-design/active-work / 0.11.0

@titan-design/active-work@0.11.0

Durable workspace state for engineering work — CLI, MCP, and Claude skill for tracking initiatives across sessions

AI Security Review

scanned 10d ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. An npm postinstall hook modifies the user's Claude Code control surface. No network endpoint was identified in this install path.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review; previous stored version diff introduced dangerous source

Trigger

Installing the package runs its postinstall lifecycle hook.

Impact

Package installation can add or replace AI-agent instructions and commands without an explicit setup command.

Mechanism

The hook replaces a Claude skill directory and writes a Claude slash-command file under the user's home directory.

Policy narrative

The package automatically runs a postinstall script. When a Claude configuration directory exists, it removes and recreates the active-work skill directory, copies bundled skill content into it, and writes an aw-prompt command into the user's Claude commands directory. This mutates a foreign AI-agent control surface during npm installation rather than through an explicit user setup action.

AI rationale

The automatic lifecycle hook makes unconsented changes to Claude Code's skill and command directories. This meets the install-control-surface blocking policy.

Evidence

package.jsonscripts/postinstall.js~/.claude/skills/active-work~/.claude/commands/aw-prompt.md

Decision evidence

Source & flagged code

10 flagged · loading source

package.jsonView file

•scripts.postinstall = node scripts/postinstall.js

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node scripts/postinstall.js

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg

dist/cli.jsView file

145Trigger-reachable persistence chain: manifest.bin -> dist/cli.js L145: } L146: const cwd = path.resolve(process.cwd()); L147: if (isInside(cwd, from)) { ... L921: frontmatter: raw.frontmatter, L922: body: raw.body, L923: prefix ... L1540: } L1541: function parseWorktreePorcelain(stdout) { L1542: const out = []; ... L2067: function claudeProjectsRoot() { L2068: const override = process.env.CLAUDE_PROJECTS_ROOT; L2069: if (override && override.length > 0) {

High

Trigger Reachable Persistence

A manifest entrypoint or package-local install chain reaches persistence behavior.

dist/cli.jsView on unpkg · L145

•matchType = normalized_sha256 matchedPackage = @titan-design/active-work@0.9.0 matchedPath = dist/cli.js matchedIdentity = npm:[redacted]:0.9.0 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/cli.jsView on unpkg

•matchType = malicious_source_fingerprint_signature signature = 44af3df91fafe989 signatureType = suspicious_hashes sourceLabel = final_verdict:malicious matchedPackage = @titan-design/active-work@0.9.0 matchedPath = dist/cli.js matchedIdentity = npm:[redacted]:0.9.0 similarity = 1.000 shingleOverlap = 6 summary = package final verdict is malicious

High

Known Malware Source Fingerprint Signature

Source fingerprint signature matches a known malicious package signature; route for source-aware review.

dist/cli.jsView on unpkg

•matchType = previous_version_dangerous_delta matchedPackage = @titan-design/active-work@0.18.1 matchedIdentity = npm:[redacted]:0.18.1 similarity = 0.500 summary = stored previous version shares package body but lacks this dangerous source file

High

Previous Version Dangerous Delta

This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.

dist/cli.jsView on unpkg

145} L146: const cwd = path.resolve(process.cwd()); L147: if (isInside(cwd, from)) { ... L921: frontmatter: raw.frontmatter, L922: body: raw.body, L923: prefix ... L1540: } L1541: function parseWorktreePorcelain(stdout) { L1542: const out = []; ... L2067: function claudeProjectsRoot() { L2068: const override = process.env.CLAUDE_PROJECTS_ROOT; L2069: if (override && override.length > 0) {

Medium

Install Persistence

Source writes installer persistence such as shell profile or service configuration.

dist/cli.jsView on unpkg · L145

scripts/postinstall.jsView file

1// Copy the bundled Claude Code skill into ~/.claude/skills/active-work/ and the L2: // /aw-prompt slash command into ~/.claude/commands/ when ~/.claude exists. Skip L3: // silently otherwise. Fail-soft: never abort an npm install if a copy fails. L4: import { existsSync, mkdirSync, cpSync, copyFileSync, rmSync } from 'node:fs'; L5: import { join, dirname } from 'node:path'; ... L12: const commandSource = join(__dirname, '..', 'claude-commands', 'aw-prompt.md'); L13: const claudeDir = join(homedir(), '.claude'); L14: const claudeSkillsDir = join(claudeDir, 'skills', 'active-work'); ... L29: } L30: mkdirSync(claudeSkillsDir, { recursive: true }); L31: cpSync(skillSource, claudeSkillsDir, { recursive: true }); L32: console.log(`active-work: installed Claude Code skill to ${claudeSkillsDi

Critical

Ai Agent Control Hijack

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

scripts/postinstall.jsView on unpkg · L1

dist/chunk-LU6VJBBO.jsView file

•stage = ast_semantic_analysis; reason = ast_alias_growth_limit_exceeded; limitedFiles = 2

High

Semantic Analysis Limited

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

dist/chunk-LU6VJBBO.jsView on unpkg

dist/aw.jsView file

•matchType = normalized_sha256 matchedPackage = @titan-design/active-work@0.9.0 matchedPath = dist/aw.js matchedIdentity = npm:[redacted]:0.9.0 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/aw.jsView on unpkg

Source & flagged code

10 flagged

package.jsonView file

Lines 27-67json

27 "engines": {

28 "node": ">=22"

29 },

30 "packageManager": "pnpm@10.28.1",

31 "scripts": {

32 "dev": "tsx src/cli.ts",

33 "build": "tsup && pnpm build:dashboard",

34 "build:cli": "tsup",

35 "build:dashboard": "vite build --config src/dashboard/vite.config.ts",

36 "gen:cli-reference": "pnpm build:cli && node scripts/gen-cli-reference.mjs",

37 "mine:session": "node tools/mine-session-signals.mjs",

38 "index:session": "node tools/build-session-index.mjs",

39 "mine:files": "node tools/mine-file-history.mjs",

40 "mine:drain": "node tools/mine-drain.mjs",

41 "mine:drain-eval": "node tools/eval-drain.mjs",

42 "export:aw-data": "node tools/export-aw-data.mjs",

43 "cost:rollup": "node tools/cost-rollup.mjs",

44 "eval:miner": "node tools/eval-miner.mjs",

45 "index:eval": "node tools/eval-session-index.mjs",

46 "check:tools": "for f in tools/*.mjs; do node --check \"$f\" || exit 1; done && node -e \"JSON.parse(require('fs').readFileSync('tools/pricing/models.json','utf8'))\"",

47 "postinstall": "node scripts/postinstall.js",

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg · L47

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg · L47

48 "preuninstall": "node scripts/preuninstall.js",

49 "test": "vitest run",

50 "test:watch": "vitest",

51 "test:unit": "vitest run --project unit",

52 "test:integration": "vitest run --project integration",

53 "test:coverage": "vitest run --coverage",

54 "typecheck": "tsc --noEmit",

55 "lint": "eslint src __tests__ --no-error-on-unmatched-pattern",

56 "format": "prettier --write \"src/**/*.ts\" \"__tests__/**/*.ts\"",

57 "format:check": "prettier --check \"src/**/*.ts\" \"__tests__/**/*.ts\""

58 },

59 "devDependencies": {

60 "@eslint/js": "^9.39.4",

61 "@types/better-sqlite3": "^7.6.13",

62 "@types/node": "^22.10.5",

63 "@types/proper-lockfile": "^4.1.4",

64 "@types/react": "^19.2.14",

65 "@types/react-dom": "^19.2.3",

66 "@vitejs/plugin-react": "^6.0.1",

67 "@vitest/coverage-v8": "^3.2.7",

dist/cli.jsView file

Lines 125-165javascript

125}

126function yearMonth() {

127 const d = /* @__PURE__ */ new Date();

128 const y = d.

scripts/postinstall.jsView file

Lines 1-21javascript

1// Copy the bundled Claude Code skill into ~/.claude/skills/active-work/ and the

2// /aw-prompt slash command into ~/.claude/commands/ when ~/.claude exists. Skip

3// silently otherwise. Fail-soft: never abort an npm install if a copy fails.

4import { existsSync, mkdirSync, cpSync, copyFileSync, rmSync } from 'node:fs';

5import { join, dirname } from 'node:path';

6import { homedir } from

dist/chunk-LU6VJBBO.jsView file

•stage = ast_semantic_analysis; reason = ast_alias_growth_limit_exceeded; limitedFiles = 2

High

Semantic Analysis Limited

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

dist/chunk-LU6VJBBO.jsView on unpkg

dist/aw.jsView file

•matchType = normalized_sha256 matchedPackage = @titan-design/active-work@0.9.0 matchedPath = dist/aw.js matchedIdentity = npm:[redacted]:0.9.0 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/aw.jsView on unpkg