LLM Mart Basic

Derive a full UI colour palette algorithmically from one or two brand colours. Darker and lighter variants for interactive states, desaturated greys from the brand hue for borders and backgrounds, and semantic colours that feel coherent with the brand rather than generic. Use whe

The persistent shell around an application — the top bar, the app launcher, the tenant and environment cue, and in heavy tools a status bar. In an estate of several applications the shell belongs to the estate rather than to any one app, and it is what carries a user between tool

Product visuals — hero shots, demos, screenshots, landing-page panels — must reproduce the real product, not a stylised poster of it. Design with real content and real data, mirror the actual output the product generates, keep every label and number truthful, and refuse fabricate

A brand's visual tone — playful or serious, rounded or angular — should be consistent across all UI elements. Shape language in typography, border-radius, and iconography communicates personality before a single word is read. Use when establishing a design system, choosing icon l

Every interactive element needs a complete set of visual states — rest, hover, active/pressed, focus, disabled, and loading. States should be derived algorithmically from the base colour, not chosen arbitrarily. Use when designing buttons, links, inputs, or any clickable componen

Choose light, dark, or combined color mode deliberately based on brand tone and user context. Offer a theme selector only when user control genuinely matters — enterprise tools, data-heavy UIs, or extended-use applications. Use when defining the base color palette, designing a de

Buttons, inputs, pills, badges, calendars, and other interactive components form a visual family — they share the same border-radius, colour logic, shadow scale, border style, and spacing rhythm. Inconsistency between them breaks the sense of a coherent product. Use when building

When data has both a tabular/list representation and a visual representation, show both simultaneously and keep them synchronized. Clicking a row highlights the corresponding element in the visual view, and vice versa. Applies to maps, diagrams, timelines, and charts that accompa

Complex data deserves multiple view modes — grid, list, table — chosen by the user based on their task. Row and item selection should use large hit areas (the whole row or card, not just a checkbox). Selected state is communicated through a subtle background colour shift. Mass ac

Orchestrator for the full dembrandt UX pipeline. Routes a UI/UX task through six ordered stages — brand foundation → design tokens → layout → components → UX polish → accessibility gate — loading the right sub-skill at each stage. Use when the task spans multiple design concerns:

Configuration UIs for domain experts — users who understand their field deeply but are not software developers — require domain language, sensible defaults, and grouping by professional concept rather than technical parameter. Use when designing settings panels, solver or algorit

Elevation — subtle shadows and layering — communicates visual hierarchy by lifting elements above the surface. Combined with border-radius, it creates the tactile quality of cards, modals, and interactive surfaces. Use when designing cards, dropdowns, modals, tooltips, or any flo

Extract a complete design system — colors, typography, spacing, components, shadows, and W3C design tokens — from any live website using Dembrandt. Runs a headless browser against the URL and returns real computed values from the DOM. Use when you need a site's actual design toke

Forms have three layers of guidance: helper text below the input explains what to enter, placeholder shows the expected format, and validation confirms correctness. Real-time validation for complex inputs. Submit enables only when the form is valid. Use when designing or reviewin

Pipeline skill — turns a URL or DESIGN.md into a concrete UI structure with decisions already made. Extracts live design tokens, normalizes them into a semantic system, applies UX principles, and outputs an actionable UI spec. Use when building UI for an existing brand from scrat

UI layout and grouping should follow Gestalt principles so users immediately understand which controls, commands, and elements belong together. Use when designing or reviewing component layout, navigation, toolbars, forms, dashboards, or any UI where visual grouping communicates

Quick global settings — currency, language, region, units — belong in a persistent, low-profile location such as a header toolbar or footer. These controls are frequent but not primary, so they use small typography and stay out of the main content hierarchy. Use when designing gl

In large applications, information architecture determines whether users can find, understand, and act on data. Naming matters. The UI should mirror the data model and signal how data can be transformed. Dangerous or irreversible changes always require a confirm dialog. Use when

A layout is not a neutral container — choosing the right layout paradigm (feed, board, table, canvas, master-detail, dashboard, gallery, timeline, map, single-focus, narrative long-scroll) is a design decision that shapes how content is understood. Landing and marketing pages get

Manage user expectations during wait times with appropriate loading states — from simple spinners to complex skeleton screens and staggered animations. Perceived performance is often more important than actual load time. Use when designing data-heavy components, handling API call

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.

{{SKILL_ENTRY:refactor}}

{{SKILL_ENTRY:release}}

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 {{CMD: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.

Daily repo hygiene — review the day's repo state, then perform the cleanups under per-action confirmation. Fast-forward only, never destructive without a yes.

Show the project's current state at a glance — stage, open tasks, open questions, overrides since the last checkpoint, current branch. Read-only.

Wire codeArbiter's statusline into ~/.claude/settings.json, or remove it.

The sanctioned task-board mutator — add a queued task, start one (flips to in-progress and stamps the date, minting a dotted ID on pick-up), or mark an in-progress task done. The only blessed write to open-tasks.md.

{{SKILL_ENTRY:security-architecture}}

{{SKILL_ENTRY:tribunal}}

Watch a PR's CI to completion — diagnose on red, notify and offer the merge on green. Never auto-merges.

Copy a file OUT of a running sandbox box to the host — host-initiated egress only (docker cp). The reverse, a host→container bind, is impossible by construction.

Tear down a sandbox box — remove its container and named volume. --keep-volume leaves the volume; with no id, prune reclaims any leaked ca.sandbox=1-labeled object. Cached images are retained.

Run a single command inside a running sandbox box and capture a JSON result — exitCode, separate stdout/stderr, and a truncated flag past the byte cap. The scriptable exec seam.

Open an interactive shell inside a running sandbox box at /work/repo. Read-only root, non-root user, no host-FS access — explore the untrusted code interactively, then exit.

Pull an untrusted repo into an ephemeral, host-FS-isolated Docker container — clone into a named volume, build a dep-cached image, run under structural isolation. Network defaults to offline. Requires Docker and nixpacks.

Vet a new or changed third-party dependency for license, provenance, and supply-chain risk before any install runs.

Inspect ADR health read-only; optionally select one ADR with --adr N.

Record user-decided ADRs or inspect their health read-only. Preserve attribution and acceptance evidence.

Okou connects to the tools your team already uses and does the work — across marketing, sales, engineering, and operations, under your control.

An open-source Digital Worker platform for reliable execution, continuous co-evolution, and building Enterprise AI assets.

A curated list of AI Agent evolution, memory systems, multi-agent architectures, and self-improvement projects. | evomap.ai

A governance harness for AI coding.

DeepSeek Harness 手机版：可直接安装的 Android APK，AI 免 Root 操作手机（Shizuku/root 可选），文件编辑只需所有文件访问权限，前台保活 + AI 通知

Local-first, governed AI agent runtime for Python — embed it in your app, or run it as a CLI or ACP server. Permissions, MCP, memory and audit replay built in.

Orbi — the factory that builds and operates AI software factories. GitHub Issues in, releases and runnable system out

Open-source AI agent harness in native Rust — GUI, CLI, headless, and webapp from one binary. Multi-provider, MCP, skills, plugins, agent teams.

The SDK for browser agents. Interact, search, extract, and fetch any site reliably across the web

Agent-native shopping for extreme value: verifiable same-product price evidence, checkout, orders, delivery, and after-sales.

The extensible power-user platform for Google Antigravity. Adds a native In-App Browser, animated Desktop Pets, revamped Gemini UI, and custom BYOK Gemini Pro k…

Open-source, single-binary, self-hosted AI agent — your models and data stay on your machine. A coding agent on par with Claude Code and a personal assistant li…

Your pocket agent. Local-first AI agents on iOS and Android — real workspaces, tool execution with approvals, and your choice of model (DSH · Claude Code · Code…

Bridge between QQ (SnowLuma OneBot v11) and DeepSeek Harness agents: social simulation, safe MCP tools, slang learning and more.

A 7×6 framework for agent architecture. 28 patterns, each placed at a coordinate, runnable Python code with verified engineering slices from Claude Code, Aider,…

针对 AI 自动化渗透 Agent 的新一代反制蜜罐，通过反向代理将API密饵载入真实业务、反向提示词注入等方式反制自动化渗透 Agent，实现多款主流通用Agent的反制上线控制。

Open-source, local-first AI agent for coding and real work. BYOK models, MCP, skills, plugins, workflows, and private knowledge bases.

A free AI-agent toolbox for Android, 一站式安卓AI Agent工具箱

Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. I…

Independent, unofficial CLI to edit CapCut and JianYing (剪映) projects — subtitles, timing, speed, volume, templates, cut long-form to shorts. No API needed, rea…