LLM Mart Basic

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Use when a Claude Code session has just compacted, is about to compact, reached context limit, resumed after /compact, or needs to rebuild its working mental model from Claude JSONL transcripts and touched files.

Use when you have a task and must decide WHO does it — yourself, a spawned subagent, or another agent in the topology that already holds the context. The rule: how much would a blank slate have to read in to do this correctly? A lot → route to a context-holding agent; a little →

Use when the system around you does not make sense yet: you just booted into a seat and do not know how the pieces fit; someone said rig, pod, seat, fleet, topology, or slice and you are not certain what they mean here; you are unsure what kind of rig you are in or what it is for

Use when a human decision, access grant, external action, or escalation must survive the operator's absence, or when an orchestrator or PM has a judgment-worthy update for the operator.

Use when starting, building, handing off, restoring, or closing a mission or slice under the default lightweight SDLC.

Use when you do not know WHERE something belongs: you learned a lesson and are unsure which file takes it; you are about to create a new doc, folder, or convention; you are asking "should this be a skill, a note, or a chain file?"; you inherited a seat and want to know what SHOUL

Use when you're operating OpenRig and need the right skill or context for fleet recovery, seat handover, new-seat orientation, a watchdog wake, cross-host reach to an agent on another machine, rig packaging, an OpenRig upgrade, systematic debugging, queue triage, or implementatio

Use when a specific `rig` command, subcommand, or flag is already known and you need its exact syntax, JSON shape, defaults, or error meaning. NOT for natural capability discovery, open-ended how-do-I questions, or choosing which OpenRig move applies.

Use when you have just been primed into an EXISTING seat through a planned handover — a different agent retired and handed you the seat plus its earned context — and you need a world model of what just happened to you. Covers how a handover differs from compaction and from a fres

Use when ending a turn, finishing a slice, blocked on another agent's work, or escalating to a human — durable work handoff via queue items so the system keeps moving across compactions, missed messages, and interruptions. Covers the hot-potato terminal-turn-rule (active work end

Use when a long-running agent may have lost the product outcome, when context was compacted, when work has crossed a major boundary, or when a fresh path-based trace is needed before the next consequential action. Not for fresh-session orientation, waking an idle seat, or checkpo

Use when you are a sitting agent near your context threshold (~85%) and a PLANNED seat transition is due — retire deliberately and hand your seat to a fresh successor primed from a packet, rather than let compaction degrade you. Covers the handover packet, the append-only lineage

Use when replacing a seat's occupant (rebuild/handover/swap), reasoning about stable-seat-identity vs fluid-occupant-identity, choosing an old-occupant disposition (retire/advise/shadow), or recording provenance for an occupant change. Two independent outcomes (continuityOutcome

Use when reasoning about what survives compaction/context-loss/restart/seat-refresh, designing a high-fidelity restore packet, or distinguishing native runtime resume vs fork vs artifact-backed mental-model rebuild. Covers the 4 failure modes that prevent honest restore (compacte

Load this FIRST, before you assume how anything here works — especially before building, modifying, or "fixing a bug" in an OpenRig or studio-box system, or when something behaves unexpectedly and you're about to treat it as a code bug. You are almost certainly NOT operating in t

When handing work to the Codex CLI earns its cost, and how to size the run: second-model review, bounded implementation hand-offs, sandbox permissions, model and reasoning effort. Use when the user asks for Codex or `codex exec`, when a change is complex or high-stakes enough tha

Write outbound promotional copy for a product or project: launch and update posts for community platforms and social media, store page descriptions and short blurbs, landing page headlines and calls to action, press-style announcements, and the naming of a product for another lan

Judgment rules for locating the correct boundary of a requested change: staying inert outside it while completing every required site inside it. Use when scope is ambiguous, a diff touches neighboring surfaces, required dependent edits are unclear, or a proposed compatibility lay

按 scarletkc 本人的自然表达习惯撰写、改写、润色和翻译文本，覆盖推文、微博、评论、聊天消息、技术观点、项目介绍、GitHub 文本（README、issue、PR、发布说明）和正式通信。当用户要求用自己的口吻写东西、把 AI 腔文字改自然、发推、回评论、点评模型或开发工具、写项目公告、写礼貌但直接的客服或正式邮件，或要求翻译时保留语气和立场，都使用本 skill，即使用户没有点名 scarletkc 或提出风格要求。

Fourteen posts of being wrong in production, compressed to checkboxes

Healthy nodes, a quiet network, 300 restarts in three days, and a latency budget measured in milliseconds

Discovery worked. Ping worked. Every TCP connection timed out, and later the tunnel only worked when someone had a terminal open.

Every VM came back. The cluster did not. Declarative systems converge on config, and the datapath isn't config.

A surprising share of AI-in-the-terminal failures aren't the AI. They're zsh, and a version of bash from 2006.

A Claude Code plugin turns standalone project configuration into a namespaced, installable extension that teams and communities can update as one unit.

None of the safety came from the model. It came from six boring habits.

Skills package instructions and references. Subagents run work in a separate context and return results. They solve different problems and can be composed deliberately.

Six hours in, one step left, everything green, and the incident that didn't happen

CLAUDE.md carries persistent project context. Skills load reusable procedures when relevant. Separating stable facts from task-specific workflows keeps both easier to maintain.

Twenty minutes recovering secrets that never existed, and the one sentence from a human that ended it

An API request routing a model's tool call through an approval gate to a remote MCP server

31 config keys, two audits, and why the first one was wrong in both directions

The official MCP Registry stores standardized server metadata rather than package code. Publishers verify a namespace, describe installation or remote access, and submit immutable versions.

Everyone looks at the Dockerfile. The file that actually leaked the key was the project file.

Remote MCP authorization uses established OAuth standards, but secure integration still requires issuer validation, least-privilege scopes, protected token handling, and server-side enforcement.

"Copy it over and switch the reference" is two steps, and the outage lives in the one nobody checks

stdio fits local processes and prototypes. Streamable HTTP fits hosted services and shared integrations. The right choice follows where the capability runs and who must reach it.

The most important rule wasn't about what I could change. It was about what I was allowed to display.

Tools perform operations, resources expose readable context, and prompts provide reusable templates. Choosing the correct primitive makes an MCP server easier to understand and govern.

Periodic multi-reviewer sweep of the whole codebase — surfaces a triaged checkpoint report.

Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose.

{{SKILL_ENTRY:post-merge-cleanup}}

Show the codeArbiter command catalog — the public command list and what each routes to.

{{SKILL_ENTRY:commit-gate}}

Stop everything and surface a rule conflict — persona vs. docs vs. code. Present both sides and the conflict-hierarchy level; the user resolves. No silent reconciliation.

{{SKILL_ENTRY:context-check}}

{{SKILL_ENTRY:context-creation}}

{{SKILL_ENTRY:debug}}

{{SKILL_ENTRY:decompose}}

Verify the active host install, package, command ownership, enforcement{{IF:pi}}, wrapper self-test, and active-dispatch coverage gap{{ELSE}}, and harmless live-fire probe{{END}}. Read-only.

Start a feature: brainstorm a spec, get it approved, then drive it test-first through the pipeline. The one entry to implementation.

Fix a confirmed bug: a failing regression test first, then a minimal fix, then the rest of the tdd gates.

Opt this repo into codeArbiter — scaffold the root-level .codearbiter/ state store.

Read-only 3-metric governance glance — override rate, small-lane rate, sprint low-confidence ratio — each with a trend arrow vs. the prior 20-commit window.

Sanctioned, logged bypass of a gate or hard rule — one audit line, then proceed.

{{SKILL_ENTRY:finishing-a-development-branch}}

Zero-onboarding, read-only dry-run of the reviewer fleet against the current uncommitted diff. Predicts reviewers, runs the state-free secret scan, writes nothing.

Trim transcript clutter to extend session lifetime — analyze, prune a copy, or toggle the after-each-turn service. Dry-run by default; gains land at resume/compaction, not the current turn.

{{SKILL_ENTRY:decision-variance}}

An open-source, privacy-first, self-hosted knowledge workspace where humans and AI agents work together 开源、隐私优先、自托管的知识工作空间，让人与智能体在此协作

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

The go-to web for your AI coding agent — local-first search, fetch, crawl & research over MCP. No API keys, no cloud, $0/query. Public beta.

Transform and optimize your markdown documentation for Large Language Models (LLMs) and RAG systems. Generate llms.txt automatically.

合乎周礼：DeepSeek-powered Zhouli-style Chinese translator, web app, and distributable Skill package.

This is a fork of the https://dockbox.dev project I made

🏆 Curated, ranked list of AI agent harnesses (100+) — plus an MCP server, llms.txt & JSON so agents can recommend them too. Rescored weekly.

A Python framework for modular, self-contained skill management for machines.

ADHD — a skill for coding agents. Tree-of-thought with pruning, built on the Claude & Codex Agent SDK. Fans out parallel divergent thoughts under different cogn…

AI Agent 驱动的开源可自部署视频工作台：将小说与剧本转为角色、场景、道具资产、分镜、视频和剪映草稿，支持跨镜头一致性、多供应商与费用追踪 | Self-hosted AI video workspace for stories, storyboards and short-form video producti…

DeepSeek Harness Desktop App: a local AI desktop workspace for DSH Sessions, projects, files, web research, plugins, and Office artifacts.

Autonomous Offensive Security, Bug Bounty & Red Teaming Agent Framework powered by Hermes Agent, specialized reasoning skills, and multi-model LLM orchestration…

⌥ Coding agent with the IDE wired in

Supercharge AI Agents, Safely

X (Twitter) Scraper API and X API Alternative. You do not need an official X developer account. You do not need to connect or use an X account for supported scr…

My AI Stand. Realtime by day, rewriting itself by night. Summon my AI superpower.

Open-source coding agent for your terminal, built in Rust and on a journey of continuous community improvement. Issues and PRs welcome.

观澜 / Guanlan：AI Agent 的中文互联网研究、阅读与信源路由工具。

Mac Agent for macOS 26: the agentic AI harness for your Mac Desktop. Computer use, automation, scripting, coding, and more. Powered by 18+ providers across loca…

Semantic version control => entity-level diffs, blame, and impact analysis on top of git. 28 languages via tree-sitter. Built for coding agents.