LLM Mart Basic

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.

Optional manual drift audit — report stale provenance-tracked docs, then per stale doc offer re-scout, re-baseline, or defer. Not the daily loop; commit-gate auto-heal owns routine maintenance.

Brownfield back-fill — scout an existing codebase and populate .codearbiter/, then lock it initialized.

Investigate-then-decide root-cause analysis for a defect whose cause is unknown. No code changes — exits to /ca-fix, /ca-adr, or a no-action close.

Greenfield decomposition interview — a layered interview that populates .codearbiter/ and locks it initialized.

Verify the active host install, package, command ownership, enforcement, wrapper self-test, and active-dispatch coverage gap. 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.

Open a pull request the only sanctioned way — clear every BLOCK-level review finding, then stage the PR. Never a direct write to the default 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.

SMARTS arbitration — reconcile architectural artifacts against the scaffold and prior decisions; every variance resolved by an explicit, user-attributed choice.

Restructure code with behavioral parity proven through unmodified pre-existing tests, then refactor. No behavior change.

Cut a release the only sanctioned way — SemVer bump from the commit log, a CHANGELOG section, an annotated tag. Takes the declared target's name as its only argument, or --dry-run to preview one with no write. The only path to a version tag.

Review a diff with the reviewer fleet, funneled to one triaged verdict. Targets the current working diff, a path, or an inbound GitHub PR.

Exploratory spike on a throwaway branch — answer a named question with disposable code. Never merges; exits to a findings note or /ca-feature.

Autonomous sprint — one interactive spec gate, then plan-to-PR execution with every auto-decision SMARTS-scored and logged. Hard gates remain true stops.

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.

`crabbox inspect` prints the full record for a single lease: state, provider,

Run named, repo-local jobs defined in your Crabbox config.

`crabbox list` shows the current Crabbox machines (leases) for a provider. It is

`crabbox login` authenticates the CLI against a coordinator, stores the

`crabbox logout` clears the stored broker token from your user config so the CLI

`crabbox logs` prints the retained command output for a recorded run.

`crabbox marketplace` previews the Crabbox credits gateway: one Crabbox billing

`crabbox media` turns a recorded desktop video into lightweight review

`crabbox open` prepares an existing SSH-capable lease for an external editor.

`crabbox pause` pauses a single lease, freeing the remote compute while

`crabbox pond` is the cross-provider peer-discovery and lifecycle surface for a

`crabbox pool` contains machine-pool helpers. `pool list` keeps the older

`crabbox ports` bridges provider-native port publishing for an existing Crabbox

`crabbox prewarm` leases a reusable box and prepares it for test runs. For

`crabbox providers` prints the provider capability matrix that the CLI compiles

`crabbox receipt <run-id>` retrieves a brokered run's committed terminal

`crabbox results` prints the structured test summary attached to a recorded

`crabbox resume` resumes a lease previously paused with [`pause`](pause.md),

`crabbox run` syncs the current dirty checkout to a box, runs a command there,

`crabbox screenshot` captures a single PNG from a desktop lease without opening a

Claude Code skills for Chinese-narration recaps from supported video files, with optional editable JianYing/CapCut draft export. | 用 Claude Code skills 为支持的视频文件…

Drive the Unity Editor from an AI agent or the terminal. The Editor serves MCP itself over HTTP, so there is no second process to run, and the isuzu-unity-cli c…

面向企业的数字员工构建与运行平台：把专业员工的经验、流程与判断标准，固化为可随时上岗、可配置、可审批、可观测的 AI 数字员工。

A local-first, Pi-powered AI agent workspace for Desktop, WebUI, and CLI

660+ muapi-hosted generative-media models plus community-submitted third-party API tools (SEO, enrichment, social, scraping) — one YAML file per entry, browsabl…

A compounding agent OS for recursive agents. Also an open source alternative to Grok Bot and Meta's Muse.

Independent desktop client for OpenCode 2. Manage projects, sessions, parallel agents, requests, and changes on Linux and macOS.

Production agent skills for Claude Code, Cursor, and any SKILL.md harness — Codex fleets, video pipeline, monorepo review bundles, multi-chain explorer.

See and manage what your coding assistants load — skills, commands, subagents, plugins and MCP servers, with real usage.

An Enterprise-Grade Full-Stack RBAC Permission Management System Built with Go + React

Create AI Agents in a No-Code Visual Builder or TypeScript SDK with full 2-way sync. For shipping AI assistants and multi-agent AI workflows.

基于 DeepSeek Harness（DSH）的稳定桌面端，集成Git、内置浏览器与记忆功能 | DeepSeek Harness desktop GUI with local workspaces, Git, browser and memory.

⚡ Control Apache Airflow with natural language via MCP. Chat with your workflows using Claude, GPT, or any LLM — no REST API calls needed. Supports Airflow 2.x…

Open-source CLI, schemas, resolver, and DSH agent tools for DSH Plugin Hub

Pi Coding Agent 中文学习蓝皮书：从安装与第一个可验收任务开始，逐步掌握 Session、Context、Skill、Extension、Subagent 与长期 Agent 工作流。

Self-hosted framework for orchestrating fleets of specialist AI agents — ensemble reasoning and a full agentic coding pipeline, model-agnostic and local-friendl…

Self-hosted, vendor-neutral control plane for your local coding agents (Claude Code & Codex). Run your agents, on any plane.

A practical AI agents handbook covering agent systems, agentic workflows, LangGraph, MCP/A2A, context engineering, agent memory, evaluation, observability, and…

Vibe Coding 从入门到精通教程｜AI 结对编程工作流｜Prompt、Skill、Workflow、上下文管理、codex实战指南

Static security scanner for LLM agents — prompt injection, MCP config auditing, taint analysis. 51 rules mapped to OWASP Agentic Top 10 (2026). Works with LangC…