Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent India travel-planning skill that uses local trip memory and guide files without evidence of hidden execution, network access, or unrelated data access.
Install only if you want India-specific travel guidance and are comfortable with the skill keeping local notes in ~/india/memory.md about trip preferences, dates, route, group type, and recommendations already given. Avoid asking it to store passport or payment details unless you intentionally want those saved locally.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
integration: pending | done | declined
## Context
<!-- First trip or repeat trip -->
<!-- Dates, route, group type, comfort level, budget notes -->
<!-- Food preferences, hygiene sensitivity, and must-do goals -->
The related skill entry presents 'hindi' for signs, menus, and quick phrases, which indicates a language-specific constraint without stating user choice or opt-in. Organizational language policy requires offering language or locale choice unless the constraint is clearly justified and documented.
The heading 'Apps — India' indicates the skill content is scoped to a specific country/locale, but the file does not state that this is an intentional region-specific guide or offer any user opt-in or alternative locale handling. Under the policy, forcing a locale without user choice or explicit justification is a natural-language policy concern.
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.
- Adding mountain roads in active monsoon without buffer days
- Treating Ladakh as a simple weekend extension
- Ignoring altitude, permits, or road-closure risk
- Booking treks without checking whether the traveler wants scenery, challenge, or comfort
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.
- Adding mountain roads in active monsoon without buffer days
- Treating Ladakh as a simple weekend extension
- Ignoring altitude, permits, or road-closure risk
- Booking treks without checking whether the traveler wants scenery, challenge, or comfort
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.
- Adding mountain roads in active monsoon without buffer days
- Treating Ladakh as a simple weekend extension
- Ignoring altitude, permits, or road-closure risk
- Booking treks without checking whether the traveler wants scenery, challenge, or comfort
The skill explicitly instructs the agent to persist user travel preferences in ~/india/memory.md without any notice, consent, or limitation on what is stored. Even though the data is travel-related rather than highly sensitive by default, dates, group composition, comfort preferences, and itinerary details can still reveal personal patterns and create a privacy risk if retained unexpectedly.
The memory behavior section directs ongoing recording of route, dates, group type, comfort level, and preferences, but does not mention that these details will be persistently stored. This increases the privacy concern because it normalizes broad accumulation of user-specific travel context over time without informed consent or clear boundaries.
The 'When to Use' section says the skill applies when a user is planning a trip to India or wants local guidance on a wide range of topics, but it does not define specific trigger phrases, constraints, or exclusion conditions. This broad natural-language scope could overlap with ordinary travel conversation and lead to unintended invocation.
The document is explicitly scoped to India-only itineraries, as indicated by the title and all route recommendations. Under the policy rule for language or locale constraints, this is a locale-specific limitation presented without any note that the user should opt into this regional focus or that the skill is intended only for India travel requests.
The title and setup instructions narrowly direct behavior toward India travel, and the trigger language suggests the assistant should proactively engage whenever India travel arises. There is no explicit statement that this locale specialization is user-selected or optional.
No suspicious patterns detected.