LLM Mart Basic

商品短视频广告。卖点 → 分镜脚本 → 分段生成 → 拼接加字幕成片。当用户说「做条广告」「短视频」「投流素材」「分镜脚本」「带货视频」时使用。

去水印去文字。带水印 / 文字 / logo 的图 → 干净图，背景纹理自然补全。当用户说「去水印」「去文字」「擦掉 logo」「把字去了」时使用。

平铺图转隐形模特立体图。平铺图 → 有体积感与版型的立体展示图。当用户说「转 3D」「立体图」「隐形模特」「把衣服撑起来」时使用。

UGC 口播种草视频。商品 + 人设 → 口播脚本与成片，达人自拍质感。当用户说「口播视频」「种草视频」「达人风格」「UGC」「真人推荐」时使用。

鞋包配饰真人穿戴图。商品图 + 模特参考图 → 真人佩戴图，落位、透视与阴影自然。当用户说「鞋包上脚」「配饰上身」「墨镜戴上」「首饰佩戴图」「包包上身」时使用。

Apply WIT (Writing Is Thinking) as a human–LLM collaborative scientific reasoning skill for scientific question formulation, finding-driven research planning, next-experiment selection, Results or Discussion review, claim–evidence and reviewer stress tests, manuscript logic audit

Use when creating skill repositories, standardizing or validating skill repo structure, setting up composer/release workflows, configuring split licensing (MIT + CC-BY-SA-4.0), fixing plugin.json / SKILL.md validation or version-parity errors, or releasing a skill version (versio

Conversational design workshop for substantial work. Interviews the human one question at a time, explores 2-3 approaches with trade-offs, and presents the design section by section for approval before writing only design.md, then stops. Combines requirements discovery with codeb

Post-implementation completion workflow for Spec-backed Plans. Use after spec-implement completes to validate, review, create stacked commits, and open a PR via code-pull-request. Triggers only with an active Spec-backed Plan after spec-implement completes, including when the use

Continue an approved Spec-backed workflow when the user says "implement", "go", "start", or "do it". After Codex Plan Mode, persist the next missing design.md or plan.json artifact and stop. When both artifacts exist, execute plan.json with TDD and report between batches.

Write approved implementation plans in one of two modes. Explicit Inline mode creates a conversational plan for bounded work. Spec-backed Plan converts an approved design.md into plan.json. Both modes stop after producing their plan output. Trigger after atelier-orchestrator sele

Skill routing and workflow orchestration. Selects Inline Plan or Spec-backed Plan, routes to the correct workflow skill, and manages transitions between phases. Use when starting any conversation or task to determine which planning mode and skill apply.

Configure a repository for Atelier's development workflow. Use only when explicitly invoked; inspect existing guidance, issue-tracker and domain-document conventions, preview the proposed configuration, and write only after approval. It does not install Atelier or initialize exte

Disciplined debugging methodology. Triggers on bug reports, test failures, "debug this", "diagnose this", unexpected behavior, build failures, integration issues, or performance regressions. Find root cause before a permanent corrective fix; contain urgent harm safely first.

Grill the user relentlessly about a plan, decision, or idea, maintaining the project's domain model (CONTEXT.md, ADRs) as decisions crystallise. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

Generate and validate conventional commit messages following the conventionalcommits.org spec. Use whenever the user wants to commit code, mentions commit messages, git commit, or asks to create a commit. Triggers on "commit", "git commit", "conventional", or when reviewing commi

Compact the current conversation into a handoff document for another agent to pick up.

Manage GitHub pull requests or GitLab merge requests: create, read/leave/respond to comments, and merge. Triggers on "open a PR", "make a PR", "merge this PR", "merge the MR", "read PR comments", "leave a comment on the PR", "respond to a comment", "ship this", or when a feature

Multi-agent code review with parallel specialized reviewers, architecture validation, challenge validation, and durable handling of previously decided findings. Use `rq` to request a review of diffs (defaults to main branch), `rs` to respond to findings and record intentional non

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.

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.

Inspect architectural variances with SMARTS; record only explicit user choices. Report-only requests make no changes.

Restructure code without changing observable behavior: rename, extract, inline, move, deduplicate, or replace an internal implementation with an equivalent one. Prove parity through unchanged pre-existing tests. Not for new behavior, bug fixes, explanation-only questions, or committing finished work.

Prepare a declared release target, or preview it with --dry-run. Derive its version and require authorization before publication.

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.

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.

Threat-model a sensitive design with STRIDE on request. Read-only analysis of threats, controls, and implementation constraints.

Run an opt-in deep codebase audit with persisted findings. Confirm cost before dispatch; filing and telemetry need separate approval.

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

Reviews a new or changed third-party dependency before adoption. The read-only

A read-only health scan of every recorded ADR under `.codearbiter/decisions/`. For each one it

Records an architectural decision as a numbered, dated ADR under `.codearbiter/decisions/`.

Assembles everything codeArbiter logs — `overrides.log`, `triage.log`, `decisions/`,

The one exception to codeArbiter's slash-command pipeline: a lightweight question-and-answer

A periodic whole-repository review. The caller supplies the reviewer unit list to

Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible capabilities.

Agent OS: the agent gets smarter on its own. We just hold the line: the grading command and expected result never make it into the success contract we hand it.…

Curated systems, benchmarks, and papers etc. on memory for LLMs/MLLMs --- long-term context, retrieval, and reasoning.

:memo: Vimlike Modal Text Editor in Rust

CI-native security testing for MCP servers. Attack simulation, schema drift detection, and health scoring before agents depend on them.

Hermes Agent memory plugin/provider for scope-aware recall, SQLite truth, LanceDB semantic search, and hybrid retrieval.

For You Agent——AI 时代的个人随身数字人格。把你的模型、AI 账号、技能、提示词和工作方式，带到每一个 AI 工具里。

A coding agent: give it a prompt and it reads, writes, runs commands, and searches code in a loop until the work is done, using native tool-calling across OpenA…

A secure persistent personal agent server in Rust. One binary, sandboxed execution, multi-provider LLMs, voice, memory, Telegram, WhatsApp, Discord, Teams, and…

Self-evolving agent: grows skill tree from 3.3K-line seed, achieving full system control with 6x less token consumption

DeepSeek Harness Desktop (dsh-desktop). EAC: Embracing All Creation (揽尽万象). Bundled Node.js runtime with full dsh-CLI kernel, one-click startup, 10 built-in UI…

See your agent think. Zero-config observability & governance for 26 AI agent runtimes: Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, Cline, Ope…

Save 94% on AI coding tokens. Index your codebase, agents search instead of reading files. Works with Claude Code, Codex, Copilot, Cursor, Gemini CLI. Local MCP…

Yet another coding agent harness, lightweight and written in go.

a coding Agent from pi. ∞ providers, sub-agents, hashline edits, and a permission gate

Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting…

🧠 Leon is your open-source personal assistant.

The Station, an open-world multi-agent environment that models a miniature scientific ecosystem.

The Frontend Stack for Agents & Generative UI. React, Angular, Mobile, Slack, and more. Makers of the AG-UI Protocol

VelaTerm = iTerm2 + Codex, The Best Terminal for AI Coding