The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch documentation and API reference material from https://docs.typesafe.ai. This involves reading external markdown files and metadata to inform the AI's actions.
[INDIRECT_PROMPT_INJECTION]: The skill provides guidance for building systems that ingest untrusted external data (referred to as "state" or "source text"). It recommends using structured JSON and specific criteria to manage this data, which creates an ingestion surface for the systems built using these patterns, although this is standard for the described use case.
[SAFE]: The skill consists entirely of instructional text and documentation links. It does not include any scripts, executable code, or persistence mechanisms. No evidence of obfuscation, credential exposure, or command execution was found.

typesafe-ai