Text-based User Interface for Sneat.app
macOS or Linux, via Homebrew:
brew install --cask sneat-co/tap/sneatOr build from source with Go:
go install github.com/sneat-co/sneat-cli/cmd/sneat@latestHuman-facing commands (auth, space/spaces, contact/contacts, chat, ui, convo) work
against Firestore directly or the sneat-go API and print a table by default (--format json|yaml|csv
to switch). The Action Protocol family below is agent-facing: every command prints JSON by default,
talks to api.sneat.cloud/v0/sneatai/* (see
sneat-ai-backend), and holds a pending action as a
local draft file (<UserConfigDir>/sneat/actions/<actionID>.json) unless --server is given.
| Command | Flags | Does |
|---|---|---|
sneat action new |
--json, --utterance, --language, --space, --server |
Start an action from a semantic candidate; validates it (client-held) or creates it on the server (--server) |
sneat action add <id> |
--json, --utterance, --language |
Apply a semantic patch — merged locally and re-validated for a client-held draft, sent to action_patch for a server-held one |
sneat action show <id> |
— | Print the local draft, or fetch a server-held action |
sneat action validate <id> |
— | Re-run validation without changing the semantic candidate |
sneat action commit <id> |
— | Commit the action; exits 2 with {"error": "..."} on stdout when the server refuses because validation still needs input |
sneat action cancel <id> |
— | Cancel the action (deletes the local draft, or calls action_cancel) |
sneat action list |
— | List local action drafts (client-held and server-held index entries) |
sneat action schema |
--offline |
Print the Action Protocol's machine-readable semantic schema (fetched by default, or compiled-in with --offline) |
sneat context |
--space |
Print a Space snapshot (contacts, lists, happenings, now/today) for an agent to read before acting |
sneat query |
--json, --space |
Ask what to buy/schedule for contacts before a horizon |
We build with our own tooling:
- SpecScore — specify requirements as
SpecScore.mdartifacts - SpecStudio — author & manage specs across their lifecycle
- inGitDB — store structured data in Git where applicable
- DALgo — data access layer for Go
- cover100.dev — drive toward 100% test coverage
- DataTug — query & explore data