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
Security audit
superstition-debunker
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local media-literacy helper that analyzes pasted claims for folklore-style patterns, with an opt-in local log that users should treat carefully.
Install is reasonable for local claim triage. Be careful with --log: pasted messages may include personal, medical, or family-chat details, and the tool stores the first 120 characters locally until the log file is removed.
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
Vulnerability Patterns
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
- 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
Findings (4)
Autonomous Decision Making
Medium
- Category
- Excessive Agency
- Confidence
- 85% confidence
- Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
- Content
md | share immediately | forwarding quotas (forward to 10...) | | act-before-you-think | "before it's too late," "hurry" | Highest weights because **deadlines disable verification — that is their function**. A message that demands speed is a message that fears checking. Suppression framing also pre-explains the absence of sources ("of course there are no news reports — they're suppressing
Autonomous Decision Making
Medium
- Category
- Excessive Agency
- Confidence
- 85% confidence
- Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
- Content
python | share immediately | forwarding quotas (forward to 10...) | | act-before-you-think | "before it's too late," "hurry" | Highest weights because **deadlines disable verification — that is their function**. A message that demands speed is a message that fears checking. Suppression framing also pre-explains the absence of sources ("of course there are no news reports — they're suppressing
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
When --log is enabled, the tool persistently stores a timestamp, verdict, hit metadata, and the first 120 characters of user-supplied claim text. That can capture sensitive or personal content from forwarded messages without any consent prompt, retention notice, minimization, or redaction, creating a privacy and data-handling risk if the logs are later accessed or shared.
- Content
Static analysis
No suspicious patterns detected.