Security audit
Security checks for vulnerabilities and agentic risk
This is a straightforward currency-conversion skill that contacts a disclosed exchange-rate API and does not show hidden persistence, credential access, or destructive behavior.
Install this if you are comfortable with a small shell-based tool that uses curl and python3 and contacts Frankfurter/ECB over the network for exchange rates. Users who need English documentation, offline conversion, or strict no-network operation should avoid or adapt it.
The skill appears intended to invoke shell-capable functionality (curl, python3, and a referenced fx-converter.sh) but does not declare any explicit tool scope such as allowed tools or permissions. This creates an authorization and review gap: the runtime may grant broader shell/environment access than readers expect, making misuse or later code changes harder to constrain and audit.
The skill's name and operational documentation are presented entirely in Chinese, with no indication that language is selectable or that the skill is intended only for a Chinese-speaking locale. This can violate language/locale policy when a skill effectively forces one language without user opt-in or documented regional justification.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
# fx-converter --list # 列出支持的货币
set -euo pipefail
API="https://api.frankfurter.app/latest"
# 货币代码 -> 中文名
declare -A CURRENCIES=(
This markdown file describes what the skill does and shows command examples, but it does not specify when the skill should activate, what user phrases should invoke it, or any exclusion conditions. In agent environments that infer activation from descriptions, the broad natural-language description of currency conversion could cause unintended invocation for general finance questions.
The script sends user-provided currency parameters to api.frankfurter.app via curl, which transmits request metadata and query contents over the network. While network access is central to the tool's purpose, the file does not include an explicit user-facing warning or disclosure beyond naming the data source in comments, and this operation is not surfaced at runtime.