Back to skill

Security audit

Generate Judgements

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent helper for generating evaluation YAML judge questions, with file reading and YAML writing tied to that purpose and gated by user confirmation.

Before installing, be aware that this skill reads the target skill directory and can write or update evaluation YAML after approval. Review any generated environment fields carefully and do not place real API keys in committed YAML files.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
- Access to the target skill directory (must contain `SKILL.md`)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 52)May include surrounding context.

md
- Access to the target skill directory (must contain `SKILL.md`)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 60)May include surrounding context.

md
- Access to the target skill directory (must contain `SKILL.md`)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger list is broad and includes generic phrases like 'write test config' and 'add judgements', which can plausibly appear in ordinary authoring conversations unrelated to this specific skill. That increases the chance of unintended skill invocation, causing the agent to switch workflows, read files, or modify YAML configs when the user did not explicitly intend to use this capability.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The guidance explicitly requires judgement questions to be written in English, which is a language policy constraint. Although it allows interacting with the user in their language, it does not offer a choice or opt-in regarding the English-only output for these generated questions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

This markdown file documents an environment section containing OPENAI_API_KEY and OPENAI_BASE_URL, which signals that users may place credentials into configuration. The surrounding text does not include any warning about secret handling, storage risks, or avoiding committing credentials to source control.

Content

No source excerpt is available for this finding.