The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the official vendor endpoint at https://app.skilder.ai/mcp for Model Context Protocol (MCP) integration. This is a standard reference to the provider's official infrastructure managed by the skill's author.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to modify tool configuration files and run CLI commands such as claude mcp add and codex mcp add. These actions are restricted to local environment configuration and target the verified vendor URL.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data (workspace roles and skills) from a remote source, creating a standard surface for indirect prompt injection.
Ingestion points: Workspace data is fetched from the https://app.skilder.ai/mcp endpoint described in SKILL.md.
Boundary markers (absent): The instructions do not specify delimiters to separate untrusted remote content from the agent's instructions.
Capability inventory: The agent is instructed to perform local configuration writes and execute CLI commands in SKILL.md.
Sanitization (absent): No explicit filtering or sanitization of the remote workspace metadata is mentioned.
[SAFE]: Authentication is handled via industry-standard OAuth 2.1 with browser consent, ensuring the AI agent does not handle or store sensitive user credentials. Additionally, the skill explicitly instructs the agent to obtain user confirmation before applying any configuration changes.

connect-to-skilder