The Agent Skills Directory

[SAFE]: The skill is a technical documentation resource for the Effect-TS ecosystem, specifically focusing on the effect/unstable/ai modules. It provides idiomatic patterns for handling streaming data from language models.
[DATA_EXPOSURE]: No sensitive file access or hardcoded credentials were found. The examples use standard logging and metrics recording patterns.
[INDIRECT_PROMPT_INJECTION]: While the skill demonstrates how to process and accumulate untrusted output from an LLM (via LanguageModel.streamText), it does so within the context of a chat history management tutorial. The examples do not demonstrate passing this untrusted data to dangerous sinks like shell commands or eval() functions.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or unverified dependency installations were detected. The imports utilize the official effect package namespace.

effect-ai-streaming