Back to skill

Security audit

JT Meeting Summary

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only meeting-summary skill with no code execution, network access, persistence, or hidden data movement, though it is strongly oriented toward Chinese-language output.

Install only if you want Chinese-oriented meeting and call summary formats. For transcripts containing phone numbers or other personal data, review the generated summary before sharing or storing it, and ask the model to omit identifiers if they are not needed.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (11)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The template and instructions require Chinese headings and specify Chinese-character limits and summary lengths, which imposes a specific language/locale on output. The file does not state that Chinese output is optional, user-selected, or justified as a region-specific skill, creating a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The mode-selection table defines user request patterns entirely in Chinese, and the hard rules require Chinese fallback strings such as 原文未涉及 and 暂无. This creates a language-policy concern because the skill appears to enforce a specific language/locale without explicit user opt-in or an offered alternative.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest describes activation as "Use $jt-meeting-summary to summarize this meeting transcript" and also enables implicit invocation. This does not define clear boundaries for when the skill should or should not activate, increasing the risk of unintended invocation from broadly matching meeting-summary requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The file mandates Chinese output tokens such as 暂无, 发言人1, and the full Chinese section headings, which effectively forces a specific language/locale. Under the policy, locale-specific behavior should either be optional for the user or clearly documented as a justified region-specific constraint.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The template instructs the model to include speaker phone numbers in the output sections even though earlier guidance says to avoid repeatedly exposing them and to mask middle digits unless raw identifiers are required. In a meeting-summary skill, this creates a real privacy leakage risk because generated summaries may propagate personally identifiable information into downstream documents, chats, or storage without any explicit warning, consent check, or stricter masking default.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The phrase "general meeting-summary requests" is a broad activation description for a markdown skill and does not define clear constraints on when this format should be used versus other summarization formats. This ambiguity can cause unintended invocation for ordinary summarization requests that may not specifically require this concise-minutes structure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The instruction 'keep it under 20 Chinese characters when possible' imposes a specific language/locale requirement. Elsewhere the schema and placeholder values are also defined only in Chinese, with no opt-in or explanation that this skill is region-specific, which matches the policy-violation criteria.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file requires summaries to be produced in Chinese characters and specifies Chinese rhetorical structure, which imposes a language/locale choice in natural-language instructions. Because no user opt-in or alternative language path is provided, this appears to violate the policy against forcing a specific language.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The example output requires a title of '不超过20字', which directs the skill toward Chinese-language output for generated titles. The file does not explain that the skill is region-specific or allow the user to choose another language, so this is a natural-language locale policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The file's suggested output is entirely in Chinese section headings and labels, which imposes a specific language on generated output. The instructions do not offer an opt-in choice or explain that the skill is intended only for Chinese-language users, so this appears to violate the language/locale policy.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The prompt hard-codes '中文为主;除原文专有英文外,禁止英文标注', which overrides normal user language preference and can force outputs into Chinese without explicit user opt-in. This is not code-execution dangerous, but it is a real prompt-policy issue because it can cause unauthorized language transformation, reduce usability, and create misleading behavior when the user expects another language.

Content

No source excerpt is available for this finding.