Read AGENTS.md before making any change. AGENTS.md is the sole repository operating guide for agents. Do not infer current release status from old branches, old PRs, old incident notes, audit documents, terminal history, or prior session handoffs. For current version and release state, use only: Repository authority: Required release order: private branch → private PR → private merge → private-main health proof → public staging → promotion guard → public PR → public merge → publish → fresh-install / E2E…
This project enforces WCAG AA accessibility standards for all web UI code. Accessibility review is enforced by three global hooks: If an edit is blocked, delegate to accessibility-agents:accessibility-lead first. Before writing or modifying any web UI code - including HTML, JSX, CSS, React components, Tailwind classes, web pages, forms, modals, or any user-facing web content - you MUST: Automatic trigger detection: If a user prompt involves creating, editing, or reviewing any file matching *.html, *.jsx, *.tsx, *.vue, *.svelte, *.astro, or…
This workspace enforces WCAG AA accessibility standards for all web UI code. Accessibility review is enforced by three global hooks, not just instructions: If the edit gate blocks you, delegate to accessibility-agents:accessibility-lead first. After the review completes, UI file edits are unlocked for the rest of the session. See docs/hooks-guide.md for the full technical breakdown. Before writing or modifying any web UI code - including HTML, JSX, CSS, React components, Tailwind classes, web pages, forms, modals, or any user-facing web…
Guidance for Claude Code (claude.ai/code) working in this repository. This is the upstream source for odoo-mcp — the Odoo MCP server published to PyPI as odoo-mcp (uvx odoo-mcp) and consumed by downstream projects as their odoo MCP server. This repo is the canonical place to fix odoo-mcp. Any bug or enhancement discovered while using odoo-mcp from a consumer project should be fixed here, not worked around downstream. Consumers run the published package — a fix only reaches them once it…
GTM Agents is an AI-powered automation platform for Go-To-Market teams. It provides 67 specialized plugins with 203 AI agents, 243 business skills, and 20+ workflow orchestrators designed for sales, marketing, growth, customer success, and revenue operations teams. Mission: Save GTM professionals 15+ hours per week on repetitive busywork by automating prospecting, content creation, analytics, and campaign orchestration. Value Proposition: Average value saved: $25,000+ per user per year. The workspace/ directory provides a consistent folder structure for all agent operations. All…
A sophisticated dual-mode assistant that facilitates collaborative knowledge sharing between Coordinators (knowledge creators) and Team members (knowledge receivers) within the Semantic Workbench platform.
Using AI Context for Development: * New developers: Read CONFIGURATION.md + PYTHONLIBRARIESCORE.md for project overview * Building assistants: - Start with ASSISTANTSOVERVIEW.md for common patterns - Use specific assistant files (e.g., ASSISTANTPROJECT.md) as implementation templates * Working on specific assistants: Load the relevant ASSISTANT_*.md file for focused context * Library development: Choose appropriate PYTHONLIBRARIES*.md file by functional area * Frontend work: Study component patterns in WORKBENCH_FRONTEND.md * API development: Follow service patterns from WORKBENCH_SERVICE.md * MCP servers: Use existing servers…
Read TODO.md at the start of each session. When asked to work on TODOs, pick the top unchecked item. After completing an item, delete its line from TODO.md. All configuration…
Integration layer between BMAD-METHOD and Ralph. bmalph bundles and installs two AI development systems: The instructions file depends on the configured platform — see src/platform/ for the mapping. npm run ci runs (in order): Ralph's shell scripts and platform drivers are tested with BATS: npm run update-bundled syncs bmad/ from the upstream BMAD-METHOD repo (tracked as a git checkout in .refs/bmad/). It pulls latest from main (or a specific ref with -- --bmad-ref <ref>), copies bmm/ and core/ into bmad/,…
This document provides context to understand the LangChain Python project and assist with development. This is a Python monorepo with multiple independently versioned packages that use uv. This monorepo uses uv for dependency management. Local development uses editable installs: [tool.uv.sources] Each package in libs/ has its own pyproject.toml and uv.lock. Suggest PR titles that follow Conventional Commits format. Refer to .github/workflows/pr_lint for allowed types and scopes. CRITICAL: Always attempt to preserve function signatures, argument positions, and names for exported/public methods.…
AI-powered biology research assistant. Inspired by NanoClaw. See README.md for setup. Single Node.js process that connects to WhatsApp, routes messages to Claude Agent SDK running in containers pre-loaded with biology…
The workflow guide diagrams are authored in draw.io (how-turboplan-connects.drawio and how-finalize-connects.drawio), each rendered to a sibling .svg. Not Mermaid. The user does visual positioning in draw.io (drag and drop); Claude handles XML-level consistency (spacing, sizing, alignment, gap normalization). Visual positioning is what draw.io excels at. Precise numerical consistency (uniform gaps, aligned coordinates, matching sizes) is tedious in the GUI but straightforward via XML edits. When the user shares a screenshot or says they moved things around, read the .drawio file…
Turbo is a modular collection of agentic coding skills with sibling editions for Claude Code and Codex. Skills connect into larger pipelines like /finalize and /review-pr. See README.md for an introduction and docs/workflows.md for workflow details and diagrams. Each docs/ guide answers one reader question: workflows.md (how it works), examples.md (what to type), requirements.md (what you need and what setup changes), customization.md (how to adapt and update). Place new doc content by the question it answers. Each skill is self-contained.…
You are working in the Codex edition tree. The default Codex instructions in this directory live in AGENTS.md, which Claude Code does not auto-load. Read it before editing. Use Codex vocabulary in codex/skills/: updateplan instead of TaskCreate, applypatch instead of Edit, ~/.agents/skills instead of ~/.claude/skills, AGENTS.md instead of CLAUDE.md, $skill-name invocation form. The full mapping is in docs/harness-vocabulary.md. Do not introduce Claude Code primitives here. When changing a skill, convention, addition, or setup/update/migration file here, check whether the parallel file…
Project: Journalism skills maintenance When a bug is reported, don't immediately attempt to fix it. Instead: Collection of Claude Code and Codex skills for journalism, media, academia, and technical workflows. This repo contains modular instruction sets that extend supported coding agents for specialized tasks. Each skill directory contains domain knowledge, workflows, templates, and best practices. Packages use .claude-plugin/plugin.json. Codex supports standards-based skill installs and a verified legacy-compatible journalism-core package route. Each stable skill has Codex UI metadata. The repository does…
Qwen3.8-Flash-Next (125B + 51B n-gram, 4-bit, ~104 GB) on Macs via SSD-streamed experts and a slot cache. Read PLAN.md for design, MEASUREMENTS.md for every measured number and its method, Tools/verify.sh for the acceptance battery. Work lands directly on main. Read the canonical engineering integration record, also projected into PLAN.md. Sevra selects and maintains a few qualified models for actual hardware profiles and optimizes the whole stack around them. The product is not permanently tied to one flagship; the current engine's…
See the repository root CLAUDE.md for project-wide rules. The binary uploaded to the cluster is the server: ssh connect uploads the local databricks CLI and runs it in server mode…
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.
Keesan12CLAUDE.md
