Instructions file CodexOpenCode
Instructions for trentisiete/endy, covering endy — multi-agent coding stack, who's in the stack, delegation primitives, personas (templates, not obligations) and cli gotchas (keep in mind).
Skill Claude CodeCodex
Use this skill when a task is best handed off to another coding agent in the endy stack — OpenCode (multi-model worker, fast refactors, test writing), CommandCode/cmd (Kimi K2.6 / DeepSeek), Hermes (Nous Research, tool-heavy), Gemini, or Claude. Covers three delivery modes: short blocking calls, long detached tmux…
Skill Claude CodeCodex
Use this skill — it is the primary one — when you need to drive another coding agent interactively through a tmux pane. Open named windows, send prompts, capture output, manage lifecycle, AND switch models / change auth / use slash commands inside the live CLI (claude, cmd, opencode, hermes, gemini, codex). Includes a…
Skill Claude CodeCodex
Use this skill when you need to inspect the endy environment of a running task — handoff lineage (who handed work to whom and why), peers in the same and other endy sessions, and per-agent tier headroom (codex 5h/wk, opencode session cost) so you can pick the next agent to hand off to. The same data is auto-prepended…
Agent
Reviews code for "taste" — does it read well, match the codebase's idioms, feel clean? Uses CommandCode's taste-1 model. NOT a bug hunter; pair with the Codex reviewer.
Agent
Mechanical refactors at scale — renames, signature changes, file splits, codemods. Fast, multi-file, model-agnostic. Best for "do this same change in 40 places".
Agent
Writes tests for existing code. Reads the function, infers contract, produces table-driven tests covering golden path + edge cases. Does not modify production code.