Kasetto
Reproduce and synchronize AI coding-agent environments from one version-controlled configuration.
- Source repo
- pivoshenko/kasetto
- Stars
- ★ 219
- Last updated
- 1d ago
- License
- NOASSERTION
- Primary language
- Rust
- FA score
- 65/100 · Some gaps
At a glance
- How it runs
- Works with
- Universal · cross-platformCodex · Claude Code
- Cost
- Free, no paid service needed
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A platform team commits kasetto.yaml and kasetto.lock so new engineers reproduce the same Claude Code, Codex, and Cursor environment.
- Not a fit if
- Teams that require a graphical environment-management interface
- Users wanting only a one-off skill install without configuration or lockfiles
- Users who need native MCP support in Pi
- Website
- kasetto.dev ↗ set on GitHub
- Maintenance
- Commits in the last 30 dayslast commit 2026-11-20latest release v3.13.1 (2026-11-20)
- Source review
- 65/100 · Some gaps
What does this agent do, and when should you use it?
Kasetto is a Rust command-line environment manager that distributes skills, commands, MCP configurations, and instruction files across AI coding agents. Users declare sources, versions, target agents, and named assets in kasetto.yaml; kst sync then fetches the content and transforms it into each agent's native directories and formats. It records the exact installation in kasetto.lock, with normal syncs honoring the lock, --update advancing versions, and --locked enforcing it in CI. Sources can reside on public or private GitHub, GitLab, Bitbucket, Codeberg, Gitea, and self-hosted instances, while configuration can be local, global, or remote. Its deployment boundary is a static CLI binary on macOS, Linux, or Windows: it is neither a hosted service nor a model runtime.
Kasetto searches for configuration through $KASETTO_CONFIG, ./kasetto.yaml, the source key in the global config.yaml, and the global kasetto.yaml, in that order. kst init scaffolds configuration; kst add and kst remove modify sources and synchronize or prune assets; kst lock resolves sources into kasetto.lock; and kst sync installs skills, commands, mcps, and instructions into target-agent directories. During synchronization it hashes content, compares the lockfile, touches only changed assets, and inserts managed blocks when instruction files must be aggregated. Configurations can compose through extends and target agents including Claude Code, Cursor, Codex, Windsurf, Copilot, and Gemini CLI in one run. For ${kst_...} placeholders, it resolves values from environment variables, credentials.yaml, or a selected external secret manager and injects them into agent settings without storing resolved secrets in the lockfile. kst list reports installed assets, kst doctor checks versions, paths, recent sync status, and broken skills, while kst clean removes tracked assets for a selected scope.
- A platform team commits kasetto.yaml and kasetto.lock so new engineers reproduce the same Claude Code, Codex, and Cursor environment.
- A developer using several coding agents distributes one set of skills, commands, and instructions with a single kst sync.
- An enterprise team publishes internal skill packs from private GitHub, GitLab, or self-hosted repositories using existing access tokens.
- A CI maintainer runs kst sync --locked to verify that the environment exactly follows the committed lockfile.
- A team sharing MCP configurations injects credentials from Vault, 1Password, or cloud secret stores through ${kst_...} placeholders.
- An organization composes organization, team, and project configurations through extends while retaining project-specific scope.
Key setup facts
Pulled from the install and usage notes below. The project README has the current values.
- Install / run
curl -fsSL kasetto.dev/install | shbrew install kasettoscoop install kasettocargo install kasetto
- Environment variables
GITHUB_TOKENGITLAB_TOKENBITBUCKET_TOKENGITEA_TOKEN- Config files
kasetto.yaml- Install notes cover
- Windows · macOS · Linux
How do you install or deploy this agent?
On macOS or Linux, use the standalone installer:
curl -fsSL kasetto.dev/install | shOn Windows, run:
powershell -ExecutionPolicy Bypass -c "irm kasetto.dev/install.ps1 | iex"Homebrew is also supported:
brew tap pivoshenko/tap
brew trust pivoshenko/tap
brew install kasettoFor Scoop on Windows:
scoop bucket add kasetto https://github.com/pivoshenko/scoop-bucket
scoop install kasettoUsers with a Rust/Cargo environment can install the crate:
cargo install kasettoThe documented public-repository path requires no additional credentials. Private repositories require the appropriate host token in an environment variable, such as GITHUB_TOKEN, GITLAB_TOKEN, BITBUCKET_TOKEN, or GITEA_TOKEN.
How do you use this agent?
Scaffold a project configuration in the current directory:
kst initAdd a source and immediately synchronize all of its skills:
kst add https://github.com/anthropics/skillsYou can instead select named asset types:
kst add https://github.com/example/repo --skill find --mcp github --command reviewRun synchronization and inspect the result:
kst sync
kst list
kst doctorA team configuration can also be synchronized directly over HTTPS:
kst sync --config https://example.com/team-skills.yamlCommit kasetto.yaml and kasetto.lock together. To advance moving references, run:
kst sync --updateEnforce the existing lockfile in CI with:
kst sync --lockedWhat are this agent's strengths and limitations?
- One YAML file declares skills, commands, MCPs, and instructions and converts them into native formats and locations for multiple agents.
- kasetto.lock separates reproducible synchronization, explicit upgrades, and locked CI enforcement.
- It supports multiple Git hosting providers, private repositories, and self-hosted instances instead of binding distribution to one agent ecosystem.
- Secrets can come from environment variables, a credentials file, or existing secret managers without being recorded in kasetto.lock.
- Static binaries cover macOS, Linux, and Windows, with JSON output and meaningful exit codes for automation.
- The documented workflow is entirely CLI- and YAML-based; no graphical configuration or review interface is described.
- Adoption requires maintaining both kasetto.yaml and kasetto.lock and understanding local, global, remote, and extends-based configuration precedence.
- Pi has no native MCP support, so Kasetto skips MCP sources for that target.
- A normal sync does not propagate a rotated secret into an existing live entry; kst sync --update is required.
- Repository metadata reports NOASSERTION for the license even though the README badge links MIT and Apache-2.0; adopters should verify the actual license files.
How does this agent compare with similar options?
The project contrasts itself with Vercel Skills and Claude Plugins. Its README characterizes Vercel Skills as installing from a curated catalog and Claude Plugins as supplying runtime integrations, while positioning Kasetto around version-controlled declarative configuration, lockfiles, and reproducible synchronization across projects, machines, and agents. This is the project's own positioning; the supplied source contains no independent benchmark.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| Kasetto This agent | 65 · Some gaps | CLIFree | ★ 219 | 1d ago | Rust | Codex · Claude Code |
| Mantishack (Mantis AI) | 57 · Major gaps | CLIFree + model costs | ★ 508 | 4mo ago | Rust | Codex |
| Agent Desktop | 81 · Good | CLIFree | ★ 26k | 3d ago | Rust | — |
| Browser4 Agentic Browser | 65 · Some gaps | CLIFree + model costs | ★ 1.2k | 5d ago | Kotlin | OpenAI API |
How does FollowAgents rate this agent?
Why each dimension lost points
The evidence shows Kasetto operates from user configuration and existing credential sources, giving permissions and secret sourcing some boundaries. The documentation says missing secrets fail, and resolved values are excluded from the lock file. AGENTS.md also instructs audit output not to echo resolved MCP definitions and describes caching only immutable refs. Deductions: sync fetches remote content and writes to agent directories, and installation instructions include piping installer scripts into shells; the supplied material does not explain per-write confirmation, general recovery procedures, or detailed dependency vulnerability management.
README, Cargo.toml, AGENTS.md, and lib.rs provide mutually supportive descriptions of configuration resolution, locking, sync, and error handling. AGENTS.md distinguishes an inability to complete a job from completing it with reported problems. Deductions: dependency requirements are generally broad version ranges, and the material provides no dependency audit results; this review did not execute the program, so runtime behavior and actual error messages remain unverified.
README lists many agents, project and global configuration, private code hosts, four asset types, and custom destinations, covering a broad set of scenarios. The skill clearly limits its purpose to reporting usage of locally installed assets and requires reporting coverage gaps and unverified formats. Deductions: the skill trigger description is very broad, and the sample does not show all agent and scenario limitations; the skill asks for pruning recommendations, which require care about coverage.
The materials provide a command table, installation methods, configuration examples, secret configuration guidance, a security policy, dual licensing, and a release workflow. The release process updates versions, and a vulnerability reporting channel is specified. Deductions: the sample does not include complete documentation, FAQ, changelog, or detailed maintenance commitments; license metadata is NOASSERTION, although repository files and Cargo.toml specify MIT/Apache-2.0 dual licensing.
Kasetto brings skills, commands, MCPs, and instructions into version-controlled configuration and a lock file, then formats them for multiple agents. The usage-reporting skill adds a clear use case by joining installed assets with local activity records. Deductions: the materials cannot establish actual sync correctness, performance, or the user’s permission and maintenance costs; pruning recommendations depend on session-store coverage.
Key README claims receive partial support from Cargo.toml, lib.rs, and the architecture descriptions in AGENTS.md. The skill also states coverage and format limits for its reported data. Deductions: performance, security, and compatibility claims remain repository assertions in several cases; the sample lacks complete source, release history, and independent validation, so confidence remains low.
- Sync fetches remote assets and writes to agent configuration directories; verify the configured sources and destination paths before use.
- Usage reports cover only agent session stores the skill can read. Incomplete coverage or unverified formats means idle results should not be treated as automatic grounds for deletion.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
Release notes excerpt
[3.13.1] - 2026-11-20 Build - justfile: Rename rust recipes to package parts - deps: Upgrade dependencies and mermaid to v12 CI/CD - Always run every part and standardize the release Release - V3.13.1View on GitHub ↗ -
Release notes excerpt
[3.13.0] - 2026-11-15 Bug fixes - subagents: Warn on lost tool arguments and refetch refs kst lock moved - sync: Keep plugin ownership safe across renames, bad input, and failed edits - justfile: Build the site before run-prod-server - site: Hide the ribbon's see-all link on narrow screens - site: Tidy the landing on phone widths Build - justfile: Add an audit recipe for known dependency vulnerabilities - justfile: Build the whole cargo workspace - Standardize just recipes - deps: Update dependencies Documentation …View on GitHub ↗ -
Release notes excerpt
[3.12.0] - 2026-10-28 Bug fixes - site: Keep the production opengraph image - site: Allow eval under the csp in development so client bundles load - site: Close the gap between the docs link and the nav rule - site: Centre the nav rule, tighten heading word gaps, ground the cassette shadow - site: Derive the docs-domain redirects from meta.json - site: Refuse to proxy a failed install script download - audit: Address review findings - audit: Keep resolved secrets out of package lookups Documentation …View on GitHub ↗
FAQ
Does Kasetto call or run an AI model?
How does authentication for private repositories work?
Are resolved secrets stored in the lockfile?
How can a team keep installations identical?
Does every target agent support every asset type?
Is Kasetto free?
Does Kasetto work on Windows?
scoop install kasetto.Is Kasetto open source?
Is Kasetto safe to use?
Is Kasetto still maintained?
Who makes Kasetto?
What is Kasetto's official website?
Thanks — we use this to improve the page.