Security audit

Giphy

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small, disclosed Giphy integration for Discord, with manageable privacy considerations around sending search keywords to Giphy.

Install this only if you are comfortable using a Giphy API key and having GIF search keywords sent to Giphy. Avoid invoking it with private, sensitive, or identifying conversation details; use concise non-sensitive search terms when possible.

A.I.G

Vulnerability Patterns

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
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Direct Prompt Extraction

High

Category: System Prompt Leakage
Confidence: 85% confidence
Finding: Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
Content: Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.
md
`https://api.giphy.com/v1/gifs/search?api_key=<KEY>&q=<ENCODED_QUERY>&limit=1&rating=g&lang=en` ## Output Rule - If a GIF URL is found: send only the URL (Discord auto-embeds). - If no result is found: send a short fallback text and ask for a better keyword.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The description says to use the skill not only when a user asks for a GIF, but also whenever a brief visual reaction "improves the flow." That trigger is broad and subjective, which increases the chance of unintended invocation during ordinary conversation.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The phrase "when the moment clearly fits" is ambiguous and relies on subjective judgment rather than specific trigger criteria. Although examples are provided, the rule does not clearly define boundaries for when the skill should not activate.
Content: No source excerpt is available for this finding.

External Transmission

Medium

Category: Data Exfiltration
Confidence: 95% confidence
Finding: This skill transmits user-derived query text to an external third party, Giphy, which creates a real data exposure boundary. If invoked proactively or on sensitive conversations, user intent, emotional state, or private topics could be sent off-platform without explicit consent, and the API key is also used in the outbound request.
Content: Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.
md
Use this endpoint shape: `https://api.giphy.com/v1/gifs/search?api_key=<KEY>&q=<ENCODED_QUERY>&limit=1&rating=g&lang=en` ## Output Rule

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The documented request template fixes lang=en, which imposes an English-language locale choice without indicating user selection or justification. This can conflict with language/locale policy expectations for multilingual users.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.