Back to skill

Security audit

Gif Picker

Security checks for vulnerabilities and agentic risk

Overview

The skill is a mostly coherent local GIF picker, but its picker can emit arbitrary local file paths from an editable index as media attachments, which creates a real local file disclosure risk.

Review this skill before installing. It is not clearly malicious, but only use it with a trusted index.json and trusted GIF_LIBRARY_INDEX/GIF_LIBRARY_DIR settings, because a modified index could make the agent attach a sensitive local file instead of a GIF. Treat populate.sh as a network-active setup script and prefer pinned, reviewed GIF sources.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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)

T09 · Insecure Skill Coding Practices

Error
Location
pick_gif.py:71
Finding

Arbitrary Local File Disclosure Through Unvalidated GIF Index Paths

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill is presented as a local GIF selector, but it also creates directories, downloads content from external sources, and populates a local library. This mismatch can mislead operators into approving a seemingly harmless media utility that actually performs network access and persistent filesystem modification.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill is presented as operating on local GIFs, but this script fetches content from external URLs and stores it locally. That expands the trust boundary, introduces a supply-chain/network dependency, and can unexpectedly import unreviewed content into the user's environment, which is inconsistent with the stated behavior of a local GIF picker.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill invokes shell commands and relies on environment-controlled paths, but it declares no explicit tool scope or permissions. That makes its real capabilities less transparent to users and policy systems, increasing the chance that network/filesystem actions occur without appropriate review or sandboxing.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
install:
      - id: setup
        kind: setup
        label: "Create GIF library directory"
        run: "mkdir -p {baseDir}/gifs"
      - id: populate
        kind: populate

Session Persistence

Medium
Category
Rogue Agent
Confidence
80% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 88)May include surrounding context.

  1. Find a stable GIF URL (GitHub raw URLs are preferred — never use Tenor)
  2. Download: curl -sL <url> -o ~/.openclaw/workspace/gif-library/gifs/<name>.gif
  3. Add entry to index.json:
    json
    "my-gesture": {
      "file": "my-gesture.gif",
    

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Adding network download capability to a skill whose purpose is selecting and sending local GIFs is unnecessary privilege and creates avoidable risk. Even if the downloads are only GIFs, remote content can change over time, tracking or policy concerns may arise, and the skill can behave in ways users and reviewers would not expect from its description.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The documentation instructs users to run populate.sh but does not clearly warn that the script fetches data from external repositories. That can cause users or agents to execute network-active shell code without understanding supply-chain and content-trust implications.

Content

No source excerpt is available for this finding.