toolang

Toolang

A description language and runtime for agents.

Tool calling turned LLMs from chatbots into agents. Toolang makes agents easy to build, run, and share.

Try Now

Start with a shared agent, no installation required:

uvx toolang run https://toolang.ai/dev.too

Or use a GitHub shorthand:

uvx toolang run briceyan/dev

Get Started

Install Toolang to build and run your own agents:

uv tool install toolang

Create a local agent:

toolang new alice

Or clone a shared agent:

toolang clone briceyan/dev bob

Extend your agents with caps — composable agent primitives:

caps alice skill add briceyan/codebase-navigation
caps alice psyche add briceyan/senior-engineer
caps alice service add briceyan/github

Caps are typically Markdown files. To create your own cap, start one locally:

caps alice skill new reviewer

Run in the foreground to watch the logs. Press Ctrl+C to stop. Use PY_LOG for more detail:

toolang run alice
PY_LOG=debug toolang run alice

Or start it in the background:

toolang start alice

In either case, open the printed WebUI link to connect to the agent:

Started agent alice: https://too.run/7001

To stop a background agent:

toolang stop alice

Common Commands

# Agents
toolang new <agent>                  # Create a local agent
toolang clone <ref> <agent>          # Clone a shared agent
toolang run <agent-or-ref>           # Run an agent in the foreground
toolang start <agent>                # Start an agent in the background
toolang stop <agent>                 # Stop a running agent

# Runtime
toolang model list                   # List available models
toolang tool list                    # List available tools

# Caps
caps [agent] psyche add <ref>        # Add a psyche
caps [agent] skill add <ref>         # Add a skill
caps [agent] service add <ref>       # Add an MCP server
caps [agent] prompt add <ref>        # Add a slash command
caps [agent] skill list              # List skills
caps [agent] list                    # List all caps

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

toolang-0.2.7.tar.gz (602.5 kB view details)

Uploaded Jun 16, 2026 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

toolang-0.2.7-py3-none-any.whl (399.6 kB view details)

Uploaded Jun 16, 2026 Python 3

File details

Details for the file toolang-0.2.7.tar.gz.

File metadata

Download URL: toolang-0.2.7.tar.gz
Upload date: Jun 16, 2026
Size: 602.5 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for toolang-0.2.7.tar.gz
Algorithm	Hash digest
SHA256	`326357b13a931038332087447a531d33a250ca8a9805abe467b5b8b7a2a61faf`
MD5	`421ab68a795e82b16e80d67f09359968`
BLAKE2b-256	`e774887ba8049fc1b495d681defc4c05df1499e4603c71736b87aa9b3cdf8468`

See more details on using hashes here.

Provenance

The following attestation bundles were made for toolang-0.2.7.tar.gz:

Publisher: publish-pypi.yml on openhat-ai/toolang

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: toolang-0.2.7.tar.gz
- Subject digest: 326357b13a931038332087447a531d33a250ca8a9805abe467b5b8b7a2a61faf
- Sigstore transparency entry: 1835609494
- Sigstore integration time: Jun 16, 2026
Source repository:
- Permalink: openhat-ai/toolang@14e6e3c11ee37d5d7ee1b0470278d13f420da6b9
- Branch / Tag: refs/tags/v0.2.7
- Owner: https://github.com/openhat-ai
- Access: private
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-pypi.yml@14e6e3c11ee37d5d7ee1b0470278d13f420da6b9
- Trigger Event: push