Skip to main content

AgentsMD

CLI tool for scaffolding and syncing AI agent workspace configuration files.

Installation

pip install agentsmd

Quick Start

# First, authenticate (get a token from https://agentsmd.live)
agentsmd login --token <your-token>

# Create a new workspace in the current folder
agentsmd init

# Push your workspace to the cloud
agentsmd sync

On another machine

agentsmd login --token <same-token>
agentsmd init                    # pick your workspace from the list

Features

  • Import-or-Generate Init — agentsmd init detects existing agent files (CLAUDE.md, .cursor/rules/, .claude/, .cursorrules, .windsurfrules, copilot-instructions.md, CONVENTIONS.md, .clinerules, .continue/, AGENTS.md, .rules, KIMI.md…) and imports them. If none are present, it asks one question (which tool?) and scans your repo (package.json scripts, framework, README) to generate a single high-quality config file.
  • Multi-Machine Sync — Push and pull workspaces across machines with a single token
  • Cross-Tool Sync — Convert between AI tool config formats (Claude Code, Cursor, Copilot, Windsurf)
  • Format Migration — Migrate existing config files between formats
  • Reconciliation — Re-derive all tool formats from a single source of truth

How workspaces work

A workspace is a named set of agent config files that lives in your AgentsMD cloud account. Each local folder is bound to one workspace, so commands like sync know what to push and pull.

You can rename or move the folder freely. The binding follows the directory. Free accounts get 1 workspace; Pro unlocks unlimited.

Commands

Command Description
agentsmd login --token <X> Authenticate via API token
agentsmd logout Clear local credentials
agentsmd init Show workspace picker (clone existing or create new)
agentsmd init --name NAME Create a new workspace (shortcut when no cloud workspaces exist)
agentsmd init --from <name> Clone an existing workspace by name
agentsmd init --directory <dir> Target directory when cloning with --from
agentsmd init --force Overwrite a target file in generate-mode if it already exists
agentsmd init --no-push Skip the automatic cloud upload at the end of init
agentsmd init --no-hide-ide Don't auto-hide .agentsmd/ from VS Code / Cursor / Zed / Sublime / JetBrains
agentsmd list List your cloud workspaces
agentsmd link <name> Bind the current folder to an existing workspace
agentsmd link --remove Remove the local binding (cloud workspace is preserved)
agentsmd delete <name> Delete a cloud workspace and all its files
agentsmd delete <name> --force Skip confirmation when deleting
agentsmd up Upload local changes to the cloud
agentsmd down Download cloud changes to local
agentsmd sync Bidirectional sync with the cloud. Default conflict policy: prompt per file
agentsmd sync --interactive Per-conflict prompt: [L]ocal / [C]loud / [M]erge (also the default)
agentsmd sync --auto Newest-mtime wins; rejected version saved under .agentsmd/conflicts/
agentsmd status Show sync status for the current workspace
agentsmd restore Undo the last sync/up/down by restoring backed-up files
agentsmd restore --dry-run Preview what restore would do without writing
agentsmd resolve <path> Redo the most recent conflict for a single file
agentsmd config get <key> Print a stored preference (e.g. conflict.strategy)
agentsmd config set <key> <val> Store a preference (e.g. conflict.strategy interactive)
agentsmd config unset <key> Remove a preference (restores default behavior)
agentsmd config list Show every stored preference
agentsmd convert --reconcile Re-derive all tool config formats from the source of truth
agentsmd convert --from <fmt> Convert between AI tool config formats
agentsmd convert --from <fmt> --to <fmt> Convert to a specific target format
agentsmd convert --from <fmt> --dry-run Print conversion output without writing files

Supported AI Tools

agentsmd init detects and syncs config from any of these. Generate-mode picks one as the target.

Tool Files
Claude Code CLAUDE.md, .claude/
Cursor .cursorrules, .cursor/rules/*.mdc
GitHub Copilot .github/copilot-instructions.md, .github/instructions/
Windsurf .windsurfrules, .windsurf/rules/
OpenCode / Codex AGENTS.md
Aider CONVENTIONS.md, .aider.conf.yml
Cline .clinerules
Continue .continue/
Zed .rules
Kimi CLI KIMI.md, .kimi/

Requirements

  • Python 3.9+

Metadata

Release files for agentsmd 1.2.4

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for agentsmd 1.2.4
File Size Uploaded Provenance Actions
agentsmd-1.2.4.tar.gz 46.8 kB Without provenance Details for agentsmd-1.2.4.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for agentsmd 1.2.4
File Interpreter ABI Platform Actions
agentsmd-1.2.4-py3-none-any.whl Python 3 none any Details for agentsmd-1.2.4-py3-none-any.whl

Total release size: 116.2 kB

Release files / agentsmd-1.2.4.tar.gz

Download URL agentsmd-1.2.4.tar.gz
Size 46.8 kB
Tags Source
SHA-256 checksum
How to use checksums
d98aed516a88de5cc0f343d6b415353feb169c52b0b4467108445d5cf4bc460c
BLAKE2b-256 checksum
How to use checksums
73c1b697e71479ecb48389ffce1a2e55bc9b3d0501d4a63086650f1c93fb3e40
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.13

Release files / agentsmd-1.2.4-py3-none-any.whl

Download URL agentsmd-1.2.4-py3-none-any.whl
Size 69.4 kB
Tags Python 3
SHA-256 checksum
How to use checksums
ac688de34edf32608df64892ca5c3b759a523d5b9138de1fccc455b46162f858
BLAKE2b-256 checksum
How to use checksums
63944113e3385a3799dad0c3b2f57ff77089c6127bf60a4c056171af4c6f8279
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.13

Security for agentsmd 1.2.4

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page