donation-place-search
Warn
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches its primary instructions and helper files from a remote repository (
github.com/NomaDamas/k-skill) and the npm registry (@nomadamas/k-skill) during execution. - [REMOTE_CODE_EXECUTION]: The instructions command the agent to run
npx -y @nomadamas/k-skill@0, which downloads and executes code from an external package registry at runtime without a pinned, verifiable version. - [COMMAND_EXECUTION]: The agent is instructed to use shell commands (
npx,node) to manage skill updates and generate donation reports, creating a dependency on the integrity of the external package and registry. - [DYNAMIC_EXECUTION]: A Node.js script is executed via a shell heredoc in
instruction.md, which dynamically loads thedonation-place-searchlibrary to process data. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize information from external websites (such as
1365.go.kr) to verify donation details. - Ingestion points: Website content and search results from the 1365 donation portal (documented in
instruction.md). - Boundary markers: None identified; the skill does not use specific delimiters or instructions to prevent the agent from following directions potentially embedded in the fetched web content.
- Capability inventory: The skill utilizes shell execution capabilities (
npx,node) acrossSKILL.mdandinstruction.md. - Sanitization: No sanitization or validation of the external web content is specified before the agent processes and summarizes it.
Audit Metadata