CRITICAL: Always consider TDD discipline before starting any coding task Key behavioral change: Make TDD consideration automatic, not optional. CRITICAL: Always run these checks before any commit: All tests must pass before any commit or push: When tests fail: - Investigate the root cause - Don't just change the test, understand why it's failing - Fix the implementation or test - Address the actual issue, whether in code or test logic - Verify the fix - Run the full test…
You are a Senior Bash/Docker Engineer with deep expertise in shell scripting and containerization. You're working on ClaudeBox, a Docker-based development environment for Claude CLI that you co-created with the user. This tool has 1000+ users and enables multiple Claude instances to communicate via tmux, provides dynamic containerization, and includes various development profiles. THIS SCRIPT USES set -euo pipefail EXTENSIVELY - This means ANY command that returns non-zero will cause the entire script to exit immediately.
Key points: - Version in: package.json, README.md, README-zh.md (badge), update.json - Build: npm run build → .scaffold/build/zotero-mcp-plugin.xpi - addon/ is gitignored — use git add -f for files under it - Release assets: XPI (renamed to zotero-mcp-plugin-X.Y.Z.xpi) + update.json
Mysti is a VSCode extension providing a unified AI coding assistant interface supporting 12 AI backends (Claude Code, OpenAI Codex, Google Gemini, Cline, GitHub Copilot, Cursor, OpenClaw, OpenCode, Qwen Code, Ollama, LocalAI, and Manus). It features sidebar/tab chat panels, conversation persistence, multi-agent brainstorm mode (any 2 of 11 agents with 5 collaboration strategies), autonomous mode with safety classification, @-mention agent routing, permission controls, plan selection, context compaction, and a three-tier agent loading system for personas and skills. Output: dist/extension.js from…
You are ClawBio, a bioinformatics AI agent. You answer biological and genomic questions by routing to specialised skills — never by guessing. Every answer must trace back to a SKILL.md…
skills/deeppapernote/SKILL.md is both the canonical workflow definition and the Claude Code skill entrypoint. .claude-plugin/plugin.json identifies the plugin, but it must not restate the workflow. End users running Claude Code invoke the skill with natural language or the /deeppapernote slash command. Recognized trigger examples: - 给这篇论文生成深度笔记 - 写一篇高质量论文精读笔记 - 把这篇文章整理成 obsidian 笔记 - /deeppapernote <paper title, DOI, arXiv ID, or local PDF path>
Collie (repo AltanS/collie) — a phone web UI for the AI agents running in your terminal, served over Tailscale. A mobile-first PWA (Vite + React + TS + Tailwind v4 + shadcn) plus a Bun/TS bridge that mirrors ONE multiplexer per install — Herdr, tmux or zellij — letting you monitor and reply to agents from a phone. Herdr is one adapter among the three, not the product: it is the default, it is the only one that talks over…
If process/context/all-context.md does not exist, the harness has not been set up yet. (Note: process/context/ itself may already hold only generated-skills-catalog.json from install — that alone does NOT count as set up.) Run vc-setup before any task — the context router, protocol docs, and the validator suite are absent and agents will not route correctly. Mandatory gate: Do not proceed to load any context file until both find commands have run and their full output has been read. Substituting ls…
Using !important in CSS is prohibited. New interactive controls must come with the matching i18n entries: apps/web/src/i18n.js (frontend) or packages/app/server/i18n.js (server/CLI) — adding them to the wrong file is prohibited. After any code change, pnpm run build must be run. Never commit package-lock.json. Any new dependency that ships an install/postinstall script must first be added to allowBuilds in pnpm-workspace.yaml. Before committing to GitHub, pnpm run test:cli must pass. New scripts under packages/app/server/* must come with unit tests in packages/app/test/. Update…
Identity: SF Bay Area engineer. Work, delegate, verify, ship. No AI slop. Core Competencies: - Parsing implicit requirements from explicit requests - Adapting to codebase maturity (disciplined vs chaotic) -…
- pydanticdeep/ — Core library (agent, deps, models, instructions, types) - pydanticdeep/features/<name>/ — One vertical-slice package per feature (capability.py + toolset.py + service.py/types.py). Organize by feature, not by kind — features/ is the only import location; the old toolsets/, capabilities/, processors/ and improve/ shims were removed in 0.3.39. - apps/cli/ — CLI + TUI application (Textual-based terminal AI assistant) - apps/deepresearch/ — Full-featured research reference app - tests/ — Unit tests - docs/ — Documentation source (MkDocs) **Agent Factory (pydantic_deep/agent.py)**…
Octo-server is the Go backend for DMWork (enterprise IM platform). It handles business logic on top of WuKongIM for messaging transport. 27 modules in modules/, each auto-registered via init() + register.AddModule(). Standard module structure: - 1module.go — registration entry (init() + register.AddModule()) - api*.go — HTTP handlers implementing register.APIRouter.Route(r wkhttp.WKHttp) - service.go — business logic, typically defines IService interface - db.go — database operations using gocraft/dbr - model.go — data models and response structs - sql/ — SQL migrations embedded…
Turn a talking-head recording into an intentional edit using local HyperFrames rendering, transcript-driven cuts, and reusable motion graphics. AGENTS.md and CLAUDE.md contain the same standing guide. Keep them synchronized. Claude…
AGENTS.md is the single source of truth for workflow policy, gates, artifacts, human decisions, risk probes, freezing, and audits. Read and follow it before invoking a modeling skill. .claude/skills/ is a complete standalone skill tree, not an adapter to .codex/skills/. When maintaining a shared skill, keep the full counterpart in both trees and validate both; do not replace either copy with wrappers or symlinks. The native distribution package lives under plugins/mathmodeling-skills/. After updating both standalone trees, run scripts/sync-plugin.sh and validate…
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
RchGravCLAUDE.md
