Ae1
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - `SKILL.md` - Complete documentation (this file)
Security audit
Security checks for vulnerabilities and agentic risk
This is a low-risk descriptive travel-language planning skill with no evidence of hidden network access, persistence, credential use, or destructive behavior.
Install only if you want informational travel communication checklists and JSON-style recommendations. Verify destination, embassy, medical, payment, and safety details with official or current sources because the skill does not fetch real-time data.
Referenced artifact was not completely inspected
- `SKILL.md` - Complete documentation (this file)
The example response is explicitly written in Chinese and recommends Chinese-language services for the traveler without indicating that language choice is optional. This creates a natural-language locale policy issue because the skill appears to assume or force a specific language context rather than offering user choice.
Without declared permissions the skill's intent is opaque and cannot be validated.
The trigger criteria are broad terms like 'language', 'communication', and 'phrases if applicable', which can match many ordinary travel or even non-travel conversations. Overbroad activation can cause the agent to invoke this skill unexpectedly, leading to prompt-surface expansion, irrelevant instruction takeover, or suppression of better-suited safeguards from other skills.
The example prompts use vague formulations like 'Help me with travel language barrier navigator' and 'Provide language framework for travel planning', which are broad enough to match common user requests. This increases accidental invocation risk and can route general conversations into a skill with its own instructions, creating unnecessary prompt-injection exposure and behavior drift.
The manifest description and module docstring frame this as a language-barrier navigation skill, but the core parsing and recommendation logic is dominated by broad travel-planning features unrelated to language barriers. While some language-specific handling exists, the actual behavior materially exceeds and dilutes the stated scope.
The manifest sets the skill language to "en", which indicates a fixed language/locale choice. Under the policy rules, forcing a specific language without user opt-in or a documented justification is a natural-language policy violation.
The manifest declares language: en, and the acceptance criteria further state "English documentation primary," which establishes an English-default language constraint. Because the skill also includes multilingual travel use cases, this may conflict with the policy against forcing a specific language without explicit user choice or justification.
The module docstring characterizes the skill generically as a travel skill, which conflicts with the manifest name/description and the response text emphasizing language-barrier navigation. This is an intent-level documentation mismatch that obscures the actual intended scope.