The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external sources, specifically email content via the Gmail API.
Ingestion points: Content is retrieved in SKILL.md (Step 6), references/node.md (Step 6), and references/frameworks.md (LangChain/ADK samples).
Boundary markers: The instructions do not define delimiters or specific system instructions to prevent the agent from following commands embedded within the fetched emails.
Capability inventory: The skill demonstrates network access (requests, fetch) and the ability to pass data to LLM agents (langchain, google-adk).
Sanitization: There is no evidence of sanitization or filtering of the retrieved email snippets before they are processed or summarized.
[EXTERNAL_DOWNLOADS]: The skill recommends installing several dependencies for its operation.
Python packages: scalekit-sdk-python (vendor SDK), python-dotenv, requests, langchain, langchain-openai, langchain-core, and google-adk.
Node.js packages: @scalekit-sdk/node (vendor SDK) and dotenv.
[COMMAND_EXECUTION]: The instructions provide shell commands for installing the necessary packages using pip and npm to set up the integration environment.

integrate-agentkit