Security audit

Aviation Regulations Search

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed aviation-regulation lookup tool that sends aviation queries to Deepsky's public API and does not show hidden access, persistence, or destructive behavior.

Before installing, be aware that aviation questions you ask through this skill may be sent to Deepsky's public API. Avoid including sensitive operational or personal information in queries, and verify returned regulatory citations against official sources for safety-critical decisions.

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

Note

Location: scripts/deepsky_search.py:49
Finding: Unbounded Remote Response Read Can Cause Memory Exhaustion
Content: View full analysis
Remediation: View remediation
MAX_RESPONSE_BYTES: raise ValueError("API response exceeds the permitted size") except ValueError as exc: if str(exc) == "API response exceeds the permitted size": raise body = resp.read(MAX_RESPONSE_BYTES + 1) if len(body) > MAX_RESPONSE_BYTES: raise ValueError("API response exceeds the permitted size") data = json.loads(body.decode("utf-8")) if not isinstance(data, dict): raise ValueError("API response must be a JSON object") matches = data.get("matches", []) if not isinstance(matches, list) or len(matches) > 20: raise ValueError("API response contains an invalid matches collection") return data ``` The error handling should also catch these validation failures and report them without printing an unbounded server-controlled response body. ]]>

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (1)

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 92% confidence
Finding: The skill instructs the agent to make unauthenticated outbound network requests to deepskyai.com but does not declare an explicit tool scope or allowed-tools policy. That creates an over-permissioning and governance gap: an agent/runtime may permit broader network access than intended, making it harder to constrain where data is sent and to audit external communication.
Content: No source excerpt is available for this finding.