Research-first Agent Skill that runs long-horizon paper pipelines with evaluator gates, heartbeat watchdogs, resume controls, and manifest cleanup.
This repository also configures its own agents. See what autoresearch-paper tells them →
Skill Codex needs its repo
Turn a paragraph-level research brief into a research-first autonomous paper pipeline. Use for the Codex-host migration with a physically separate, plan-bound Claude Code/MiniMax M3 worker session, hash-bound evaluator evidence, authenticated lifecycle actions, typed failures, pause/resume/stop, patrol, and owned…
Skill Codex needs its repo
Run the Codex-hosted AutoResearch MVP-0 P1–P6 loop: compile and freeze a falsifiable Research IR, dispatch a fixed Claude Code/MiniMax Worker, record evidence, decide the Gate, recompile bounded successors, and supervise the run with a three-layer L0/L1/L2 Watchdog. Use for MVP-0 research-loop tests and bounded…