developing-with-streamlit
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documents the installation of several community-maintained Python packages for extending Streamlit functionality, such as
streamlit-aggrid,streamlit-folium, andpygwalker. It includes a 'Use with caution' advisory section recommending that developers verify the maintenance and popularity of these third-party components before adoption. - [REMOTE_CODE_EXECUTION]: In
skills/using-streamlit-cli/SKILL.md, the documentation describes thestreamlit runcommand's ability to execute scripts directly from a remote URL (e.g., GitHub). This is a native feature of the Streamlit CLI being documented for legitimate development use. - [COMMAND_EXECUTION]: The skill provides instructions for using the
streamlitCLI and theuvpackage manager for environment setup and application execution. These are standard development workflows for the Streamlit framework. - [DATA_EXPOSURE]: The skill specifically addresses security in
skills/connecting-streamlit-to-snowflake/SKILL.mdby recommending the use of.streamlit/secrets.tomlfor credential management and providing instructions to ensure this file is added to.gitignoreto prevent accidental exposure. - [PROMPT_INJECTION]: Documentation for building chat interfaces in
skills/building-streamlit-chat-ui/SKILL.mddescribes standard patterns for handling user input in LLM-powered applications. While this creates a surface for indirect prompt injection, the skill itself is a template for UI construction and does not contain malicious instructions.
Audit Metadata