Security audit

Startup Pipeline

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent startup-research purpose, but it tells the agent to send idea text to a third-party API and update memory after each candidate without clear user consent or retention limits.

Install only if you are comfortable with startup idea details potentially being sent to idea-reality-mcp.onrender.com and retained in agent memory. Before use, require the agent to ask before any external API submission, send only a redacted summary, and store only user-approved, sourced candidate notes with an expiration or deletion path.

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 (2)

other

Warning

Location: SKILL.md:119
Finding: Uncontrolled Disclosure of Startup Idea Data to a Third-Party API
Content: View full analysis
Remediation: View remediation

T02 · Agent Memory Poisoning

Warning

Location: SKILL.md:263
Finding: Unbounded Persistent Memory Updates from Candidate Research
Content: View full analysis
Remediation: View remediation

SkillSpector

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)

Ae4

Medium

Category: analysis-evasion
Confidence: 80% confidence
Finding: Suspicious Unicode normalization or mixed-script content
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill is written as Russian-first operational guidance and does not offer a user language preference or opt-in, which can cause the agent to respond in an unintended language and degrade user comprehension. This is primarily a safety/usability issue rather than a direct security exploit, but it can increase the chance of misunderstandings in a workflow that includes payments, legal compliance, and infrastructure decisions.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.