LIghtJUNction

Use this skill when working with the lightflow.control_if workflow that selects between two JSON values.

Use this skill when working with the lightflow.control_merge workflow that merges two JSON values.

Use this skill when working with the lightflow.control_split workflow that splits JSON arrays or objects.

Use this skill when working with the lightflow.control_switch workflow that selects JSON values by key.

Use this skill when working with the lightflow.image_crop workflow that crops PNG image artifacts.

Use this skill when working with the lightflow.image_edit workflow that creates deterministic preview image edits.

Use this skill when working with the lightflow.image_inpaint workflow that creates deterministic preview inpaint images.

Use this skill when working with the lightflow.image_invert standard workflow, inverting PNG outputs, or composing image inversion after another LightFlow image workflow.

Generate and synchronize the canonical daily work report index. Use when updating WORK_REPORT/index.md, adding diary/report navigation, maintaining collapsible archives, or checking that daily work report links are current.

Bilibili social workflow for the assistant. Use when reading Bilibili comments, posting Bilibili comments or replies, or reading/sending Bilibili private messages with the user's cookie-backed session.

Low-star open-source contribution workflow. Use this when preparing or doing small, low-risk PRs/issues, candidate queues, or maintainer follow-up for early-stage active repositories.

Low-star open-source contribution workflow. Use this skill when exploring small, recently active GitHub repositories, preparing local candidate queues, doing tiny docs fixes, typo fixes, broken-link fixes, small tests, obvious low-risk bugs, light code review, or following up on previously opened PRs.

AgentMail mailbox workflow for the assistant email lightjunction@agentmail.to. Use when reading, sending, configuring, or documenting the assistant's AgentMail email address, mailbox identity, CLI usage, or email safety rules.

Automatic cleanup workflow for Kimaki/OpenCode operations. Use when archiving completed sessions, reducing Discord thread noise, cleaning local workspace scratch areas, pruning temporary files, checking stale local state, or closing no-op scheduled runs safely.

Use for scheduled or user-triggered maintenance of the owner's own GitHub projects: triaging notifications, reading issues/mentions, checking Actions and releases, making small fixes, and leaving clean handoff notes.

Meta workflow for improving the assistant's own skills. Use when repeated work, recurring mistakes, copied prompts, repeated command sequences, or stable operating rules should be compressed into a reusable skill instead of being rediscovered in every session.

MCP control skill for Oxygen Not Included. Defines the control loop (sense → analyze → decide → act → verify), tool calling strategies, parameter selection, batch patterns, and error recovery. Use when controlling ONI through MCP tools.

当用户要求测试、验证或审计 ONI MCP 服务器功能、连接状态、工具可用性，或评估 MCP 操作体验和易用性时使用。指导 agent 执行系统化的功能测试流程，覆盖连接、查询、地图、建造、相机、订单等核心能力，并生成测试报告。

当开始新的 Oxygen Not Included 殖民地，或用户要求开局设置/bootstrap 配置时使用。涵盖思考前暂停、错峰日程、按属性重命名复制人并由用户确认命名风格、禁用自动消毒、侦察起始星体，以及规划早期双侧扩张和打印舱实验室。

当用户要求 agent 通过 MCP 循环游玩 Oxygen Not Included、自动玩一段时间、继续殖民地，或运行暂停-规划-恢复循环时使用。强制执行严格的 pause -> observe -> plan -> execute -> resume briefly -> pause -> verify 循环，读取 game_control domain=ui uiDomain=edit_mark action=list，限制运行窗口，并在风险或歧义决策前停下等待用户确认。

当用户要求检查 ONI 复制人状态、怀疑复制人卡住/被困/不可达、报告复制人在错误地点发呆或睡觉、要求救援分诊，或需要快速复制人健康/导航审计时使用。指导 agent 先用 dupes_control domain=info action=status_check，只检查被标记区域，除非用户明确授权，否则不要下达救援动作。

当用户要求快速总结 ONI 附近环境/资源、开局区域概览、周围资源、附近危险、扩张方向，或问“我周围有什么”时使用。指导 agent 使用低 token 的只读 MCP 工具，尤其是 read_control domain=world action=area_snapshot，并返回简洁、可执行的环境/资源摘要，不执行游戏动作。

当用户询问 ONI 游戏机制、建筑/资源/元素/动植物用途、策略建议，或在未授权控制游戏时问“我该做什么”时使用。事实性答案优先依据可验证的外部文档或静态仓库资料；仅当建议依赖当前存档时读取只读殖民地、世界和资源状态，不调用已禁用的游戏内 knowledge/database/guide 查询。

当用户要求快速演示、展示 ONI MCP 能力、功能介绍、capability showcase、试玩 MCP 工具、看看 MCP 能做什么，或说出"演示一下""快速展示""showcase"时使用。执行一条预定义的 10 步快速演示流程，覆盖虚拟文件、视图切换、复制人管理、日程、消毒、截图、工具流程发现、实时世界分析、建造挖掘预检和规划功能。目标是快速、低确认、直观地向用户展示 MCP 工具链能力。

Use this skill when working with the lightflow.flux_image_edit workflow, prompt-guided image editing, or FLUX image-edit model sync.

Use this skill when working with the lightflow.flux_inpaint workflow, masks, local repainting, or FLUX inpaint model sync.

Use this skill when working with the lightflow.flux_preview_text_to_image deterministic preview workflow or testing FLUX pipelines without a model backend.

Use this skill when working with the lightflow.flux_text_to_image workflow, syncing FLUX models, or generating images from prompts.

Use this skill when working with the lightflow.flux_text_to_image_router workflow or routing image generation between real FLUX and preview runtimes.

Use this skill when working in the CortexFS repository and the user asks to test CortexFS, test FUSE, test /ctx, test ctx, test agent.sh, run live tests, use Ollama or smollm2:135m, verify file/socket ABI behavior, check tests/mounts/cortexfs, inspect the systemd mount, update the AUR package, or diagnose mount/socket permission failures.

This skill should be used when the user asks to "build a CortexFS tool", "create a CortexFS agent", "package a CortexFS extension", "write a cortexfs.object/v1 or cortexfs.object/v2 manifest", "install an external tool", or "install an external agent".

Benchmark live CortexFS multi-agent system with Inspect AI ctx CLI. Use when asked test ctx agents, compare architect/coder/reviewer/worker performance, measure quality, latency, TTFT, throughput, tokens or errors, diagnose benchmark regressions, produce repeatable agent evaluation report.

在 CortexFS 仓库中继续去重、减少重复逻辑、合并等价代码、减少代码量或函数数量、恢复上次 DRY 重构现场，或在新增代码前避免写出重复 helper 时使用。

This skill should be used when working with douyin hidden commands, undocumented command surfaces, cookie auth commands, root compatibility crawling, Obscura integration commands, or when the user asks about commands not shown in the normal CLI help.

This skill should be used when working with this repository's douyin-cli package and douyin command-line tool: installing it with uv tool install, running official Douyin OpenAPI commands, managing OAuth auth, validating packaging with uv_build, or troubleshooting CLI-only behavior.

This skill should be used when the user asks to develop AndroidUse modules, Zygisk payloads, dynamic hook tools, AndroidUse MCP tool registration, Rust .so modules, or says "开发androiduse模块", "动态hook", "动态mcp工具注册", "zygisk loader", or "payload.so".

This skill should be used when the user asks to connect Codex to the AndroidUse Android MCP server, retrieve the AndroidUse MCP secret from an adb-connected device, refresh the secret with action.sh, configure ANDROIDUSE_MCP_SECRET, or set up adb forward for the androiduse MCP server.

Use this skill when creating, running, or modifying the lightflow.video_auto_edit_plan workflow for automated video editing plans.

Use this skill when rendering, testing, or modifying the lightflow.video_render_edit workflow and its ffmpeg-backed demo renderer.

Debug and verify MagicNet on a real Android root device. Use when working on MagicNet MCP usage, adb/root validation, eBPF or netd status, DNS leak diagnosis, sing-box/TUN behavior, certificate-less packet capture, eCapture/tcpdump checks, support bundles, device logs, or module runtime issues under /data/adb/modules/MagicNet.

Use when working in the human-mcp repository: MCP endpoint behavior, ask_human tool semantics, Rust backend, Bun web UI submodule, AUR packaging, or Arch deployment.

Use this skill when working with the lightflow.rig_llm workflow, routing prompts through Rig to OpenAI-compatible APIs, Anthropic, Ollama, OpenRouter, DeepSeek, xAI, or local test providers.

This skill should be used when working on the dependency-free agent.sh CLI as a CortexFS frontend, including AI calls through the filesystem ABI, MCP stdio, and cluster task submission.