gemini-cli

One example I ran into recently is asking Gemini CLI to do something that isn't possible: use multiple tokens in a Gemini CLI custom command (https://github.com/google-gemini/gemini-cli/blob/main/docs/c...). It pretended it was possible and came up with a nonsense .toml defining multiple arguments in a way it invented so it couldn't be read, even after multiple rounds of "that doesn't work, Gemini can't load this." So in any situation where something can't actually be done my assumption is that it's just going to hallucinate a solution. Has been good for busywork that I know how to do but want to save time on. When I'm directing it, it works well. When I'm asking it to direct me, it's gonna lead me off a cliff if I let it.

From https://github.com/google-gemini/gemini-cli/pull/5342#issuec... : > Would .ipynb format solve for this? Unfortunately there's not yet a markdown format that includes output cells (likely due to the unusability of base64 encoded binary data). There are existing issues TODO to create a new format for Jupyter notebooks; which have notebook-level metadata, cell-level metadata, input cells, and output cells. API facades like OpenLLM and model routers like OpenRouter have standard interfaces for many or most LLM inputs and outputs. Tools like Promptfoo, ChainForge, and LocalAI also all have abstractions over many models. What are the open standards for representing LLM inputs, and outputs? W3C PROV has prov:Entity, prov:Activity, and prov:Agent for modeling AI provenance: who or what did what when. LLM evals could be represented in W3C EAR

Apparently that ship has sailed. Claude Code and Gemini CLI requires Node.js installation, and Gemini README reads as if npm is a tool that everybody knows and has already installed. https://www.anthropic.com/claude-code https://github.com/google-gemini/gemini-cli