Back to skill

Security audit

Prediction

Security checks for vulnerabilities and agentic risk

Overview

This forecasting skill is coherent and disclosed: it stores local prediction notes for its stated purpose and does not show hidden execution, credential use, or exfiltration.

Before installing, understand that the skill is designed to keep forecasting notes under ~/prediction/ across sessions. Avoid storing sensitive personal, medical, legal, financial credential, or unreleased company details there, and choose explicit-only activation if you do not want forecasting behavior to appear proactively.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The invocation guidance says to use the skill whenever a user needs a forecast and lists very broad domains including business, technical, policy, sports, market, and personal planning. It does not provide explicit trigger phrases, exclusions, or negative examples, which makes activation scope ambiguous for a markdown skill description.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation guidance is broad enough to trigger on generic questions like 'what is likely' or 'what will happen,' which can cause the skill to activate unexpectedly and influence interactions outside the user's clear intent. In a forecasting skill, this creates integrity and consent risks by routing conversations into persistent forecasting workflows without sufficiently explicit user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The setup directs the agent to create local files and persist user forecasting data, but it does not require a clear user-facing disclosure or consent step before storage begins. This is dangerous because users may unknowingly have preferences, assumptions, open forecasts, and shorthand examples retained across sessions, creating privacy and data retention risks.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · setup.md (reported line 44)May include surrounding context.

touch ~/prediction/scorecard.md touch ~/prediction/reference-classes.md touch ~/prediction/assumptions.md chmod 700 ~/prediction ~/prediction/archive chmod 600 ~/prediction/memory.md ~/prediction/forecast-log.md ~/prediction/scorecard.md ~/prediction/reference-classes.md ~/prediction/assumptions.md

text

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · setup.md (reported line 45)May include surrounding context.

touch ~/prediction/reference-classes.md touch ~/prediction/assumptions.md chmod 700 ~/prediction ~/prediction/archive chmod 600 ~/prediction/memory.md ~/prediction/forecast-log.md ~/prediction/scorecard.md ~/prediction/reference-classes.md ~/prediction/assumptions.md

text

If the files are empty, initialize them from `memory-template.md`.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The template directs creation of persistent memory files under ~/prediction/, which can cause user forecasting data, preferences, and potentially sensitive notes to be stored on disk without any explicit notice, consent flow, retention limits, or storage safeguards. In this skill, the stored fields include durable lessons, blind spots, assumptions, and forecast logs, so even if the intent is organizational, it creates a privacy and data persistence risk if users do not expect long-term local storage.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.