The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the official Anthropic API (api.anthropic.com) and the @anthropic-ai/claude-code package from the npm registry. These are well-known, trusted sources for interacting with Claude models.
[COMMAND_EXECUTION]: The skill uses standard shell commands to verify the presence of required tools (which claude) and includes a pattern for parsing JSON responses from the Anthropic API using a Python one-liner via a shell pipe. These commands are used to process data from official, trusted endpoints.
[PROMPT_INJECTION]: The skill contains templates for orchestrating sub-agents by interpolating task descriptions and context into prompts. This behavior is consistent with the primary purpose of an AI orchestration bridge; while it represents a surface for indirect prompt injection, it does not include instructions intended to bypass safety filters.

bridge-claude