Back to skill

Security audit

kakao-local-cli

Security checks for vulnerabilities and agentic risk

Overview

This skill appears purpose-aligned for a Kakao Local API CLI, but users are directed to install executable code from an unpinned external repository that was not included for review.

Review the external GitHub repository before installing, prefer a pinned release or commit, use pipx or an isolated environment, avoid running installation as root or administrator, and treat any address, coordinate, or search query as data sent to Kakao's API.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:15
Finding

Installation from an Unpinned External Repository

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 15–17
Vulnerability Type: Supply-chain risk from a mutable, unverified source
Risk Level: Medium

Complete Source Snippet:

text
설치
- pipx install . 또는 pip install . (레포 클론 후)
- 저장소: https://github.com/ChloePark85/kakao-local-cli

Technical Analysis

The documentation instructs users to clone an external GitHub repository and install its current contents through pip or pipx. It does not specify an immutable commit hash, a versioned release artifact, cryptographic hashes, or signature verification. The project being audited contains only SKILL.md, so the package source and installation behavior cannot be reviewed locally.

Because the repository reference is mutable, the software installed by a user may differ from the content that existed at audit time. Python installation can execute attacker-controlled build backend or packaging logic. If the repository, maintainer account, or release process is compromised, malicious code could consequently execute during installation or later CLI invocation.

This finding establishes an insecure supply-chain installation path; it does not establish that the referenced repository currently contains malicious code.

Attack Path

  1. An attacker compromises the referenced repository, its maintainer account, or another mechanism controlling its default branch.
  2. The attacker adds malicious package source, build-system logic, or installation hooks.
  3. A user follows the documented instructions and clones the repository without selecting a reviewed immutable revision.
  4. The user executes pip install . or pipx install ..
  5. The Python packaging process builds or installs the attacker-controlled content.
  6. Malicious code executes with the privileges of the installing user, either during installation or when kakao-local is invoked.
  7. The code could access data available to that user, includi ...[truncated 723 chars]
Remediation
View remediation

Remediation Suggestions

  1. Include the complete, reviewable CLI source and dependency manifests in the skill package rather than relying solely on a mutable external repository.
  2. If external retrieval remains necessary, pin installation instructions to an immutable, reviewed commit hash or a specific release artifact.
  3. Publish cryptographic hashes for release artifacts and require users or installation automation to verify them before installation.
  4. Sign releases or commits and document signature verification using a trusted maintainer identity.
  5. Pin Python dependencies to reviewed versions and use hash-locked dependency files, such as a requirements file containing --hash entries.
  6. Use an isolated virtual environment or pipx environment and explicitly advise against installation with administrator or root privileges.
  7. Audit packaging files and build-system configuration for executable installation hooks before publishing each release.
  8. Add the reviewed version or commit identifier directly to SKILL.md so users cannot unintentionally install a later, unaudited repository state.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill documentation states that it is suitable for agents and automation pipelines, but it does not clearly warn that user-provided queries, addresses, and coordinates are transmitted to the external Kakao Local API. This can lead to unintended disclosure of sensitive location or address data by downstream users or agents who assume processing is local-only.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The natural-language content of the skill description and usage guidance is presented in Korean, and the file does not indicate that this language choice is optional or region-specific. Under the language/locale policy, forcing a specific language without user opt-in can be a policy concern.

Content

No source excerpt is available for this finding.