awaken

Library Listed

Rust AI agent backend: build capabilities once, tune behavior live, serve every client

Checked · Sources

Downloads
172
Dependents
0
Stars
100
Forks
10
Contributors
2
Commits
3,835
Latest release
0.6.0 ·

About

From the project README ·

Awaken Agents

This repository hosts Awaken Agents and its awaken-runtime execution core. The runtime protocol/specification and conformance surface are licensed under Apache-2.0; code packages may use their own file or package license metadata.

Install

Release archives contain the awaken executable, its matching awaken-sandbox execution companion, this README, the Docker Compose quickstart, and the Apache 2.0 license. The awaken executable already contains the web console; Node.js is not required at runtime.

Linux x86-64 and macOS (Apple Silicon or Intel) can use the same installer. It selects the release target, verifies the archive SHA-256, validates the binary's exact version, and installs to ${XDG_BIN_HOME:-$HOME/.local/bin} without sudo:

curl -fsSL https://github.com/awakenworks/awaken/releases/download/v1.0.0/install.sh | sh -s -- v1.0.0
awaken --version

Set AWAKEN_INSTALL_DIR to choose another destination. For provenance-sensitive environments, download install.sh and install.sh.sha256, verify the script, then run it; every release asset also has a GitHub build-provenance attestation.

Windows x86-64 can use the checksum-verifying PowerShell installer:

Read the full README
$installer = Join-Path $env:TEMP 'install-awaken.ps1'
Invoke-WebRequest https://github.com/awakenworks/awaken/releases/download/v1.0.0/install.ps1 -OutFile $installer
powershell.exe -NoProfile -ExecutionPolicy Bypass -File $installer v1.0.0
awaken --version

For provenance-sensitive Windows installations, verify install.ps1 against install.ps1.sha256 before running it. The full platform mapping is:

SystemRelease target
Linux x86-64x86_64-unknown-linux-gnu
macOS Apple Siliconaarch64-apple-darwin
macOS Intelx86_64-apple-darwin
Windows x86-64x86_64-pc-windows-msvc

Each archive has a sibling .sha256 file. Verify it before extracting the archive. GitHub also publishes a build-provenance attestation for every release asset; with the GitHub CLI installed, verify it with gh attestation verify <archive> --repo awakenworks/awaken.

To build from source, install Rust 1.96.0, Node.js, and pnpm 11.6.0, then run:

pnpm install --frozen-lockfile
cargo build --locked --release -p awaken-cli --bin awaken
./target/release/awaken --version

Try Awaken

Start the installed program:

awaken

Running awaken without a command is the same as awaken all-in-one: it starts the API, embedded web console, and local worker, persists all local state under ~/.awaken, and opens http://127.0.0.1:8080. Use --port, --data-dir, or --no-browser for common local overrides. The resulting binary contains the complete console and needs no external web directory or Node.js at runtime.

The default execution adapter follows the host's implemented capability: Linux uses bubblewrap Namespace isolation, macOS uses Seatbelt Namespace isolation, and Windows uses the host-backed Local adapter because Awaken has no Windows Namespace adapter. Linux therefore requires a usable bwrap; run awaken doctor for an actionable readiness result. Windows Local execution and an explicitly configured Local tier have no OS isolation and are appropriate only for trusted, single-user workstations.

To deliberately use programs already installed on the host on any platform, set the one deployment authority in ~/.awaken/config.toml (on Windows, %USERPROFILE%\.awaken\config.toml):

sandbox_tier = "local"

A Worker whose primary tier remains Namespace/Container can additionally opt in to exact host execution with sandbox_trusted_host_enabled = true. In both cases the Environment must bind a trusted_host + workdir execution policy; the setting only advertises Worker capability and never silently changes an isolated Environment.

The Local adapter launches ACP CLIs and shell commands as host child processes with the Session workspace as their working directory. Files, repositories, MemoryStores, Skills, and outputs still enter through Agent resource bindings; prompts receive only provider-relative paths and no second resource authority is created. The child is nevertheless unconfined and can access anything the host user can access, so TrustedHost is never a cloud or multi-tenant boundary. Use an MCP server when an existing local service should be called through a narrower capability instead of ambient filesystem access.

The Console guides one canonical path:

  1. Connect a model provider or sign in to a detected ACP runtime.
  2. Choose a starter and define one real task.
  3. Review and publish the immutable Agent revision.
  4. Follow the durable Session and its committed events.

Run the same product from the signed container image when a local binary is not convenient:

docker compose -f deploy/compose.yaml up -d
docker compose -f deploy/compose.yaml logs awaken

Open http://127.0.0.1:8080 and use the one-time setup token printed in the logs. The named volume preserves local state across container replacement. This Compose preset is for single-user evaluation: the Management container is the outer isolation boundary and its embedded Worker uses the local sandbox tier. Use the operator configuration below for shared or multi-tenant systems.

Before startup, diagnose the selected configuration, data directory, and listener without creating product state. ACP discovery remains available under the same diagnostic command:

awaken doctor
awaken doctor --json
awaken doctor acp

Build with Awaken

After the first Session is running, open API & protocols in the Console for copyable Managed Agents, Vercel AI SDK, AG-UI, A2A, and MCP examples. Every protocol uses the same published Agent and durable Session authority.

Runtime-internal examples under crates/sdk/devtools/awaken-runtime-examples/ teach embedding and extension mechanics; they are not a second product quickstart.

Operate Awaken

All-in-one, split Control/Coordinator/Worker topology, typed configuration, database migration, secret projection, and private service boundaries are owned by deploy/README.md. The k3d material is a distributed verification topology, not the local onboarding path.

For implementation boundaries, runtime coverage, architecture invariants, and documentation checks, continue to docs/README.md. To build, test and commit changes, start with CONTRIBUTING.md.

License

Apache License, Version 2.0. See LICENSE.

Runs on
GPU support unknown
What shows it
  • no dependency or feature matches a backend row
Native code
Optional
  • Binds native code: libsqlite3-sys (feature sqlite)
Python bindings
No Python binding crate found
License
Apache-2.0 OR MIT

Compare with similar projects

Order: strongest tier first, then most downloads, then name, with unmeasured values last. A library counts downloads over 30 days. An app counts binary downloads per release. An app with none counts its crate downloads over 30 days. ML Rust runs this catalogue, so its closest project always gets a row.

awaken beside the projects that share a topic with it
ProjectDownloadsDependentsStarsLatest releaseRuns on (GPU)Native codeLicense
awaken This page Listed

Rust AI agent backend: build capabilities once, tune behavior live, serve every client

172 downloads over 30 days 0 100 0.6.0 GPU support unknown Binds native code: libsqlite3-sys (feature sqlite) Apache-2.0 OR MIT

Core protocol types and traits for the Agent Client Protocol

941,949 downloads over 30 days 145 245 5.0.0 GPU support unknown No native code found Apache-2.0
adk-rust Listed

Rust Agent Development Kit - Build AI agents in Rust with modular components for models, tools, memory, and more.

5,419 downloads over 30 days 25 743 2.3.0 CUDA, Metal Binds native code: alsa-sys (feature desktop-audio), audiopus_sys (feature openai-webrtc), cxx (feature livekit), espeak-rs-sys (feature kokoro), libsqlite3-sys (feature sqlite) and 4 more Apache-2.0

A high-performance, type-safe framework for building multi-agent workflow systems in Rust

5,308 downloads over 30 days 8 388 0.8.1 GPU support unknown No native code found MIT
chidori Listed

The agent framework where every run is durable, replayable, and resumable by default. Plain async TypeScript agents on an embedded pure-Rust JS engine.

Python bindings

257 downloads over 30 days 0 1,372 3.8.1 GPU support unknown Binds native code: aws-lc-rs, aws-lc-sys, lcms2-sys, libffi-sys, libsqlite3-sys and 4 more Apache-2.0
opencrabs App Listed

The autonomous, self-improving AI agent. Single Rust binary. Every channel. Recommended: the 40MB prebuilt binary for macOS, Linux and Windows: https://github.com/opencrabs/opencrabs/releases

125 binary downloads per release Not applicable 1,027 0.5.15 GPU support unknown Binds native code: libsqlite3-sys, llama-cpp-sys-2, opusic-sys (feature local-tts), tree-sitter (feature code-graph), alsa-sys (feature local-stt) and 1 more MIT
axp App Listed

An open protocol that gives AI agents a secure, sandboxed place to run code and use tools — with lean context usage. AXP works alongside MCP

Not measured Not applicable 3 Not measured Not measured Not measured Apache-2.0
Why ListedNo archive, New or Featured rule places it, so it is Listed.

What it lacks for Featured

  • It has 0 active dependents and 0 new dependents in 365 days. Featured needs 11 active dependents or 9 new dependents in 365 days. The repository route reads 100 stars, 10 forks and 2 contributors. It needs 4,539 stars, 434 forks and 79 contributors together.

What else the check found

  • The last activity is on 8 Nov 2026, 1 day ago. The rules expect an update within 365 days. Dormancy rule

Exact rules: Listed, Featured library. The check on applied thresholds version 9.

Every rule outcome, as the check records it
  • expected-interval: version 0.6.0 base 140, raised to the 365-day floor: E = 365 days
  • dormancy: last activity 2026-11-08, 1 days ago, E = 365: active
  • featured: not featured: active_users 0 < 11 and new_adopters 0 < 9 and repository route stars 100 (needs 4539), forks 10 (needs 434), contributors 2 (needs 79)
  • listed: listed: no archive, new or featured rule applies
Crates18 crates

Measurements and sources

Each signal row shows the raw value, its source and the fetch date. The Sources list gathers every distinct source on this page. Movement needs more history.

Sources6 sources
Adoption0 active dependents, 0 new dependents in 365 days1 not measured
Adoption measurements
SignalValueSourceFetchedCounts
Active dependents0crates.io database dumpYes
New dependents (365 days)0crates.io database dumpYes
Peak active dependents0crates.io database dumpYes
Active dependents as a share of the peak1crates.io database dumpYes
Binary downloads per releaseNot measuredYes
MaintenanceLatest human commit: 1 day ago3 not measured
Maintenance measurements
SignalValueSourceFetchedCounts
Latest human commit2026-11-08GitHub repositoryYes
Latest release2026-06-08crates.io database dumpYes
First release2026-04-03crates.io database dumpYes
Latest version0.6.0crates.io database dumpYes
Days the oldest outside issue waits for a reply236GitHub repositoryYes
Maintainer replies over 365 days0GitHub repositoryYes
Repository archivedfalseGitHub repositoryYes
Repository gonefalseGitHub repositoryYes
Release commit or manifest in the repotruecrates.io packageYes
Published docstruecrates.io database dumpYes
Latest release fails cargo checkNot measuredYes
Build check dateNot measuredYes
Build check toolchainNot measuredYes
SecurityUnpatched vulnerabilities: none
Security measurements
SignalValueSourceFetchedCounts
Unpatched vulnerabilities past the grace windownoneRustSec advisory databaseYes
RustSec unmaintained advisorynoneRustSec advisory databaseYes
Deprecation noticenonecrates.io database dumpYes
Deprecated requirementnonecrates.io database dumpYes
Distribution3 releases, 172 downloads over 30 days5 not measured
Distribution measurements
SignalValueSourceFetchedCounts
Releases3crates.io database dumpYes
Downloads over 30 days172crates.io database dumpYes
Binary downloads over the last 3 releasesNot measuredYes
Homebrew installs on request over 365 daysNot measuredYes
Repology repository familiesNot measuredYes
Docker pull growth over 90 daysNot measuredYes
Docker pulls at the growth baselineNot measuredYes
Every version a prereleasefalsecrates.io database dumpYes
Every version 0.0.xfalsecrates.io database dumpYes
Every version yankedfalsecrates.io database dumpYes
Crate deleted from crates.iofalsecrates.io database dumpYes
Repository100 stars, 10 forks, 2 contributors, 3,835 commits. Stars alone never qualify a project.
Repository measurements
SignalValueSourceFetchedCounts
GitHub stars100GitHub repositoryYes — one part of the repository route
GitHub forks10GitHub repositoryYes — one part of the repository route
Contributors with a GitHub account2GitHub contributorsYes — one part of the repository route
Commits on the default branch3835GitHub repositoryNo
Repository's main crate left out of the entrynonecrates.io database dumpYes
Facts that never count
Facts that never count measurements
SignalValueSourceFetchedCounts
LicenseApache-2.0 OR MITcrates.io database dumpNo
Runs on, from listed crates and their direct dependenciesGPU support unknowncrates.io database dumpNo
What shows each backendno dependency or feature matches a backend rowcrates.io database dumpNo
Native code, from listed crates and their dependenciesBinds native code: libsqlite3-sys (feature sqlite)crates.io database dumpNo
Python bindings, from direct dependencies of listed cratesNo Python binding crate foundcrates.io database dumpNo
Crates that depend on a listed crate directly0crates.io database dumpNo

Corrections and review history

Report a correction

No correction applies to this entry.

Judgments

A judgment enters only through an approved issue and a merged pull request.

No judgment applies to this entry.