Security audit

Research Assistant Deep

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed web research workflow with a small credibility helper, and the main risk is that the helper can overrate deceptive domain names.

Reasonable to install for research assistance, but do not treat its credibility script as authoritative. Review source domains yourself, especially for lookalike hostnames, and prefer exact-domain or official-source verification for important claims.

A.I.G

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

Warning

Location: scripts/credibility.py:15
Finding: Trusted domains are validated using unsafe substring matching
Content: View full analysis
Remediation: View remediation
bool: host = host.rstrip(".").lower() trusted_domain = trusted_domain.rstrip(".").lower() return host == trusted_domain or host.endswith("." + trusted_domain) def score(url: str) -> dict: parsed = urlparse(url) if parsed.scheme not in {"http", "https"} or not parsed.hostname: raise ValueError("A valid HTTP or HTTPS URL is required") domain = parsed.hostname.encode("idna").decode("ascii").rstrip(".").lower() high_domains = { "wikipedia.org", "arxiv.org", "nature.com", "sciencedirect.com", } if any(is_domain_or_subdomain(domain, trusted) for trusted in high_domains): tier = "high" ``` This accepts legitimate hosts such as `en.wikipedia.org` while rejecting deceptive hosts such as `wikipedia.org.attacker.example` and `fake-wikipedia.org`. ]]>

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 94% confidence
Finding: The declared description presents a broad deep-research assistant capable of multi-step investigation, source comparison, synthesis, and citation tracking. The supplied code does none of that: it neither fetches web content nor analyzes multiple sources, and it has no reporting or research workflow. Its actual function is narrowly limited to heuristic classification of a URL's credibility tier from the domain name. That is a materially different primary purpose, so this is a clear mismatch.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.