The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill utilizes tools to fetch external documentation from a remote API, creating a surface for indirect prompt injection where malicious content in the documentation could influence agent behavior.
Ingestion points: Tools such as bitrix-search, bitrix-method-details, and bitrix-article-details fetch data from https://mcp-dev.bitrix24.com/mcp.
Boundary markers: The instructions do not specify boundary markers or instructions to isolate the external documentation from the agent's core logic.
Capability inventory: The skill empowers the agent to write REST calls, build applications, and manage platform deployments based on the fetched data.
Sanitization: There is no mention of sanitization or verification processes for the content retrieved from the external documentation server.
[COMMAND_EXECUTION]: The skill provides instructional shell commands for users to configure Model Context Protocol (MCP) servers in their local development environment.
Evidence: Commands for adding servers to Claude and Codex clients using claude mcp add and codex mcp add.
[EXTERNAL_DOWNLOADS]: The skill directs users to an official starter repository for development templates.
Evidence: References the github.com/bitrix24/b24-ai-starter repository for Nuxt 3, PHP, Python, and Node.js templates.

bitrix24-ai-tools