Loading...
Loading...
Found 186 Skills
Use this skill when the user wants to interact with remote Sprites from their local machine — listing sprites, executing commands, managing checkpoints, transferring files, controlling network policy, or coordinating work across multiple sprites.
Execute an approved implementation plan in a separate session with checkpoint reviews. Use after writing-plans when the user wants batched progress updates before more work continues.
Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.
Build comprehensive, mobile-compatible Obsidian study vaults from academic course materials with checkpoint-based workflow, error pattern recognition, and quality assurance. Battle-tested patterns from 828KB/37-file projects. Works across all subjects - CS, medicine, business, self-study.
Orchestrate the full paper pipeline end-to-end. Manage state propagation between phases (literature → plan → code → experiments → figures → tables → writing → review), support checkpointing and resumption. Use for assembling a complete paper from components.
Migrates Nango syncs from deleteRecordsFromPreviousExecutions()/trackDeletes to trackDeletesStart/trackDeletesEnd for automated deletion detection (including checkpoint-based full refresh). Use when updating existing createSync code.
[Tooling & Meta] Restore workflow context from checkpoint after session loss
Manage long development sessions with structured progress tracking. Creates SESSION.md files for multi-session handoff, checkpoints progress with WIP commits, and captures learnings to CLAUDE.md. Trigger with 'start session', 'checkpoint', 'wrap session', 'resume session', or 'context getting full'.
Structured checkpoint format for requesting human input. When an agent needs a decision, it must stop, present context, show options, and wait. Activate when delegating to subagents, running background tasks, or hitting any decision point that requires human judgment.
Orchestrates multi-day execution of complex tasks through milestones. Each milestone goes through plan-crafting, run-plan (worker-validator), and review-work phases with checkpoint/recovery. Triggers when the user says "long run", "start long run", "execute milestones", or "run all milestones".
Run this repo’s Units+Checkpoints research pipelines end-to-end (survey/综述/review/调研/教程/系统综述/审稿), with workspaces + checkpoints. **Trigger**: run pipeline, kickoff, 继续执行, 自动跑, 写一篇, survey/综述/review/调研/教程/系统综述/审稿. **Use when**: 用户希望端到端跑流程（创建 `workspaces/<name>/`、生成/执行 `UNITS.csv`、遇到 HUMAN checkpoint 停下等待）。 **Skip if**: 用户明确要手工逐条执行（用 `unit-executor`），或你不应自动推进到 prose 阶段。 **Network**: depends on selected pipeline (arXiv/PDF/citation verification may need network; offline import supported where available). **Guardrail**: 必须尊重 checkpoints（无 Approve 不写 prose）；遇到 HUMAN 单元必须停下等待；禁止在 repo root 创建 workspace 工件。
Automatically creates semantic Git checkpoint commits during AI coding sessions. Replaces opaque platform checkpoints with transparent, queryable Git commits using Conventional Commits format with Git Trailers. You MUST follow this skill whenever you make code changes — commit after each meaningful edit.