Security audit

yapp

Security checks for vulnerabilities and agentic risk

Overview

This skill fits a voice-journal assistant purpose, but it automatically polls private recordings and stores personal information from transcripts without clear consent, review, or deletion controls.

Install only if you are comfortable giving the skill access to private Yapp transcripts and having selected personal details retained by the agent. Use a dedicated Yapp API key if possible, confirm where it is stored, and periodically review and delete any memories created from recordings.

A.I.G

Vulnerability Patterns

Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
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

Findings (1)

T02 · Agent Memory Poisoning

Warning

Location: SKILL.md:25
Finding: Untrusted Voice Transcripts Can Poison Persistent Agent Memory
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill instructs the agent to store facts, preferences, and commitments extracted from private voice journals, but it does not warn the user that sensitive personal information may be persistently retained. This creates a meaningful privacy and consent risk because highly intimate data from recordings can be captured into long-term memory without explicit, informed authorization.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The markdown defines activation as when the user says "check my yapps" "or similar," but does not specify what similar phrases are in scope or provide exclusions. That ambiguity can cause unintended invocation from everyday utterances that resemble the example phrase.
Content: No source excerpt is available for this finding.

Ssd 3

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: Persisting personal facts, preferences, and commitments from private voice journals creates a direct data retention and secondary-use risk, especially because journals often contain sensitive health, relationship, financial, or emotional information. Once stored in agent memory, those details may be surfaced in unrelated contexts, over-retained, or exposed through other memory access paths.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.