Back to skill

Security audit

Faktum Android

Security checks for vulnerabilities and agentic risk

Overview

This Android app skill is not malicious, but it should be reviewed because it requires broad private local context reads and can change Android skill/tooling state without clear scoping or version pinning.

Install only if you want this skill tied to that exact Faktum local environment. Before using it, require the agent to list and get approval for each external path, skip unrelated vault or memory content, pin Android CLI skills and updates to reviewed versions, and avoid distributing updated skills across agent directories unless you explicitly intend that.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:10
Finding

Mandatory Access to Private Vault and Persistent Agent-Memory Files

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:10-14
Additional Location: SKILL.md:57
Vulnerability Type: Excessive access to private project context and persistent Agent memory
Risk Level: Medium

Vulnerable Code Snippet:

markdown
1. `~/Documents/ntriq-vault/wiki/decisions/2026-04-19-android-cli-채택.md` — 도입 결정/스택
2. `~/Documents/ntriq-vault/wiki/tools/android-cli.md` — CLI 사용법·커맨드 레퍼런스
3. `~/Documents/ntriq-vault/wiki/decisions/2026-04-18-faktum-news-첫-사업-확정.md` — Faktum 사업 컨텍스트
4. `~/Documents/ntriq-vault/wiki/system/faktum-news-current-state.md` — 현재 웹 스택/DB/API 스냅샷
5. `~/.claude/projects/-Users-youareplan/memory/project-faktum-news.md` — 프로젝트 메모리 + 모바일 섹션

The workflow also requires access to application source outside the Skill package:

markdown
5. Faktum 웹 API 계약 Read (`~/Documents/Projects/faktum-news/app/api/`)

Technical Analysis

The Skill requires the Agent to read files outside both the Skill package and the prospective Android project. These files include a synchronized private vault, internal business and architecture records, persistent Claude project memory, and application API source.

This violates least-privilege principles because access is mandatory and is not constrained by explicit user approval, file-content filtering, or a minimum-data requirement. Persistent Agent-memory files may contain historical instructions, user-specific information, or unrelated project context. Loading that information into the active session expands the amount of sensitive data available to all subsequent instructions and tools.

There is no evidence that the Skill transmits the accessed information to an external party. The risk is therefore unauthorized or excessive local information access rather than confirmed data exfiltration.

Attack Path

  1. A user invokes the Skill through one of its configured Android or Faktum-related trigger phrases.
  2. The Agent follow ...[truncated 1098 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove the requirement to read the persistent Claude memory file by default.
  2. Ask for explicit user consent before accessing any path outside the active project.
  3. Replace broad mandatory reads with a list of optional, task-specific context files.
  4. Request a minimal API schema or selected endpoint files instead of reading the entire API directory.
  5. Use project-relative paths or user-approved configuration rather than hardcoded, user-specific absolute paths.
  6. State which information is required from each file and prohibit use of unrelated content.
  7. Add a pre-access summary that lists every external path and allows the user to approve or reject each one.
  8. Ensure sensitive content loaded from these files is not copied into generated source, logs, or external requests unless separately authorized.

T08 · Insecure Dependencies

Warning
Location
SKILL.md:50
Finding

Unpinned Dynamic Skill Installation and Preview CLI Updates

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:50-52
Additional Location: SKILL.md:76-77
Vulnerability Type: Mutable and unverified supply-chain components
Risk Level: Medium

Vulnerable Code Snippet:

bash
# 스킬 추가 (시점 도래 시)
android skills add --skill play-billing-library-version-upgrade  # 구독
android skills add --skill r8-analyzer                            # 크기 최적화

The Skill also recommends recurring updates to a preview CLI and states that installed skills are updated dynamically:

markdown
- Google 공식 스킬(`android-cli` 등)은 `android skills add`로 업데이트됨 → 수정 금지. 우리 보강사항은 **이 래퍼 스킬에만** 기록.
- CLI는 preview(v0.7). 브레이킹 체인지 가능. `android update` 정기 실행.

Technical Analysis

The installation commands do not specify immutable versions, revisions, checksums, signatures, or an explicitly validated repository. The recurring android update recommendation likewise permits future CLI behavior to change after this Skill has been audited.

Labeling components as official does not provide technical verification of package identity or integrity. If the package registry, update channel, upstream account, name-resolution mechanism, or distribution infrastructure is compromised, the same legitimate-looking commands may retrieve altered components.

The risk is amplified by the statement that Android CLI skills are distributed to multiple Agent directories. A compromised component could consequently influence more than one Agent environment. The audited project does not itself contain a malicious dependency or payload, so this finding concerns unsafe dependency acquisition rather than confirmed malicious-code execution.

Attack Path

  1. An attacker compromises an upstream skill, registry entry, update channel, publisher account, or package-resolution mechanism.
  2. A malicious or altered release is published under a skill name used by this workflow.
  3. The Agent executes an unpinned `andr ...[truncated 1113 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin every installed skill and CLI component to an exact version or immutable revision.
  2. Record the authoritative source, publisher identity, and expected package identifier.
  3. Verify cryptographic signatures or published checksums before activation.
  4. Require manual review and approval for updates rather than recommending unconditional recurring updates.
  5. Test upgrades in an isolated environment before distributing them to active Agent directories.
  6. Maintain a lockfile or equivalent manifest containing approved versions and integrity metadata.
  7. Disable automatic distribution to Agent environments that do not require the component.
  8. Document rollback procedures and retain the last verified component versions.
  9. Treat preview CLI releases as untrusted changes until their command behavior and installation scripts have been reviewed.
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger list includes broad everyday phrases such as '모바일 앱' and '앱 만들자', which can cause the skill to activate in conversations that do not specifically intend to use this Android workflow. Overbroad activation increases the risk of unintended execution of its instructions, including unnecessary reading of local files and project memory.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest text states that the skill is triggered by Korean phrases and does not indicate that other languages are accepted or that the Korean-only constraint is optional. This creates a natural-language locale restriction without user opt-in or an explicit documented justification.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill directs the agent to read multiple broad local knowledge sources, including a general vault and cross-agent project memory, before beginning Android app work. This expands data access beyond the minimum necessary for scaffolding and implementation, increasing the chance of unnecessary exposure of sensitive local context and unintended propagation of unrelated information into the session.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The manifest describes composing official Android skills to create a Faktum Android app, but these lines add capabilities to install additional skills for billing and binary-size analysis. Those toolchain-extension actions are not necessary for the core claimed purpose and broaden the skill into package/tool maintenance.

Content

No source excerpt is available for this finding.