san-tian

Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.

Spawn parallel background subagents to handle multiple independent tasks concurrently. Each subagent works autonomously and delivers results back to the user.

Long-running iterative development loops with pacing control and verifiable progress. Use when tasks require multiple iterations, many discrete steps, or periodic reflection with clear checkpoints; avoid for simple one-shot tasks or quick fixes.

Long-running iterative development loops with pacing control and verifiable progress. Use when tasks require multiple iterations, many discrete steps, or periodic reflection with clear checkpoints; avoid for simple one-shot tasks or quick fixes.

Creates and converses with Codoxear-visible, Codoxear-managed Codex/Pi sessions through the local Codoxear HTTP API. Use when the user asks to create, start, spawn, resume, send messages to, chat with, rename, mark, or manage a session that should appear in Codoxear/Nova, especially web-owned or tmux-backed Codex/Pi sessions.

Unified GLM-5.1 LoRA deployment skill for Tinker adapters: resolve adapter from PFS or GPU Lease HTTP archive, merge into BF16, export repaired FineGrainedFP8, auto-detect SGLang or vLLM, and serve through the selected backend.

Long-running iterative development loops with pacing control and verifiable progress. Use when tasks require multiple iterations, many discrete steps, or periodic reflection with clear checkpoints; avoid for simple one-shot tasks or quick fixes.