You can now access many Seed Hypermedia features from the command line.

Overview

The Seed CLI is a command-line interface for interacting with the Seed Hypermedia network. It allows you to fetch documents, search content, manage accounts, and perform various operations on the hypermedia network.

Seed servers are open source and self-hostable servers running on the peer-to-peer Hypermedia network. You can use the CLI to interact with any Seed server using its API.

Browse the open source Seed Hypermedia repository here: https://github.com/seed-hypermedia/seed

Learn more at https://seed.hyper.media

Download the Seed desktop app: https://download.seed.hyper.media

Installation

You can run the Seed CLI directly using npx without installation:

npx @seed-hypermedia/cli@latest

If you want to install globally, you can install via npm:

npm i -g @seed-hypermedia/cli

Now you have access to the seed CLI:

seed -—help

Global Options

-V, --version: Output the version number
-s, --server <url>: Server URL (default: "https://hyper.media")
--json: JSON output (default)
--yaml: YAML output
--pretty: Pretty formatted output
-q, --quiet: Minimal output
-h, --help: Display help for command

Commands

Activity

`activity` - List activity events

View recent activity across the network.

Options:

-l, --limit <n>: Page size
-t, --token <token>: Page token for pagination
--authors <uids>: Filter by author UIDs (comma-separated)
--resource <id>: Filter by resource
-q, --quiet: Output summary only

Document Operations

`document get <id>` - Fetch a document, comment, or entity

Fetch content by Hypermedia ID.

Options:

-m, --metadata: Fetch metadata only
--md: Output as markdown
--frontmatter: Include YAML frontmatter (with --md)
-r, --resolve: Resolve embeds, mentions, and queries (with --md)
-q, --quiet: Output minimal info

Example:

seed document get --md "hm://HYPERMEDIA_URL_HERE"

`document create <account>` - Create a new document from markdown

Publishes a new document under the specified account from markdown content.

seed document create <account> --path /my-document --body-file content.md --key <keyname>

The command creates three signed blobs: a genesis change (the document's identity anchor), a document change (containing the actual content operations), and a ref (linking the document to a path and account). These are pushed to the server automatically.

Options:

-p, --path <path>: Document path (e.g. "my-document"). Auto-generated from title if omitted.
--title <title>: Document title (overrides H1 from markdown).
--body <text>: Markdown content inline.
--body-file <file>: Read markdown content from a file.
-k, --key <name>: Signing key name or account ID.

Example:

seed document create z6MkAccountUID --path /blog/my-first-post --body-file post.md --key mykey

`document update <id>` - Update document metadata or append content

Modifies an existing document's title, summary, or appends new content blocks.

seed document update <hm-id> --title "New Title" --key <keyname>
seed document update <hm-id> --body-file new-section.md --key <keyname>
seed document update <hm-id> --title "Title" --summary "Summary" --key <keyname>

The command fetches the current document state, computes the change DAG depth, creates a new signed Change + Ref pair referencing the current heads as dependencies, and submits them to the server.

Options:

--title <title>: Set document title.
--summary <summary>: Set document summary.
--body <text>: Markdown content to append inline.
--body-file <file>: Read markdown content to append from file.
--parent <blockId>: Parent block ID for new content (default: document root).
--delete-blocks <ids>: Comma-separated block IDs to delete.
-k, --key <name>: Signing key name or account ID.

`document changes <targetId>` - List document change history

View the change history for a specific document.

Options:

-q, --quiet: Output CIDs and authors only

`document stats <id>` - Get interaction statistics for a document

Get metrics and statistics for a specific document.

`document cid <cid>` - Fetch raw IPFS block by CID

Retrieve raw IPFS content by Content Identifier.

Search and Query

`search <query>` - Search for documents

Search across the hypermedia network.

Options:

-a, --account <uid>: Limit search to specific account
-q, --quiet: Output IDs and titles only

`query <space>` - List documents in a space

Query documents within a specific space.

Options:

-p, --path <path>: Path prefix
-m, --mode <mode>: Query mode: Children or AllDescendants (default: "Children")
-l, --limit <n>: Limit results
--sort <term>: Sort by: Path, Title, CreateTime, UpdateTime, DisplayTime
--reverse: Reverse sort order
-q, --quiet: Output IDs and names only

`children <space>` - List child documents

Shorthand for query --mode Children.

Options:

-p, --path <path>: Path prefix
-l, --limit <n>: Limit results
-q, --quiet: Output IDs and names only

`citations <id>` - List documents citing this resource

Find documents that reference or cite a specific resource.

Options:

-q, --quiet: Output source IDs only

Account Management

`account get <uid>` - Get account information

Retrieve information about a specific account.

Options:

-q, --quiet: Output ID only

`account list` - List all known accounts

List all accounts known to the system.

Options:

-q, --quiet: Output IDs and names only

`account contacts <uid>` - List contacts for an account

Get the contact list for a specific account.

Options:

-q, --quiet: Output names only

`account capabilities <id>` - List access control capabilities

View access control permissions for a resource.

Comments and Discussions

`comment get <id>` - Get a single comment by ID

Retrieve a specific comment.

`comment list <targetId>` - List comments on a document

Get all comments for a specific document.

Options:

-q, --quiet: Output IDs and authors only

`comment create <targetId>` - Create a comment on a document

Posts a comment on any document, with optional reply threading.

seed comment create <target-hm-id> --body "My comment" --key <keyname>
seed comment create <target-hm-id> --file comment.md --key <keyname>
seed comment create <target-hm-id> --body "Reply" --reply <comment-id> --key <keyname>

Options:

--body <text>: Comment text inline.
--file <path>: Read comment text from a file.
--reply <id>: Reply to an existing comment by its ID.
-k, --key <name>: Signing key name or account ID.

`comment discussions <targetId>` - List threaded discussions on a document

Get threaded discussion view for a document.

Options:

-c, --comment <id>: Filter to specific thread
-q, --quiet: Minimal output

Key Management

`key` - Manage signing keys

Manage cryptographic keys for signing and authentication.

Subcommands:

`key generate` - Generate a new signing key

Generate a new signing key from a BIP-39 mnemonic.

Options:

-n, --name <name>: Name for the key (default: "default")
-w, --words <count>: Mnemonic word count: 12 or 24 (default: "12")
--passphrase <pass>: Optional passphrase for key derivation
--show-mnemonic: Display the mnemonic (DANGER: write it down securely)

`key import <mnemonic>` - Import a key from existing mnemonic

Import a signing key from an existing BIP-39 mnemonic phrase.

Options:

-n, --name <name>: Name for the key (default: "imported")
--passphrase <pass>: Optional passphrase for key derivation

`key list` - List stored signing keys

Display all stored signing keys.

`key show [nameOrId]` - Show key information

Display information about a specific key or the default key.

Options:

--show-mnemonic: Display the mnemonic (DANGER)

`key remove <nameOrId>` - Remove a stored key

Permanently delete a stored signing key.

Options:

-f, --force: Skip confirmation prompt

`key default [nameOrId]` - Set or show default signing key

Set the default key for signing operations, or show the current default.

`key derive <mnemonic>` - Derive account ID from mnemonic

Derive and display the account ID from a mnemonic without storing it.

Options:

--passphrase <pass>: Optional passphrase for key derivation

Configuration

`config` - Manage CLI configuration

Configure CLI settings.

Options:

--server <url>: Set default server URL
--show: Show current configuration

Common Use Cases

Reading a Document as Markdown

seed document get --md "hm://document-id-here"

Reading with Full Context Resolution

seed document get --md --resolve --frontmatter "hm://document-id-here"

Searching for Content

seed search "your search terms"

Listing Documents in a Space

seed query "space-id" --sort UpdateTime --reverse

Getting Account Information

seed account get "account-uid"

Viewing Comments

seed comment list "document-id"

Managing Keys

# Generate a new key and show the mnemonic
seed key generate --name mykey --show-mnemonic

# Import an existing mnemonic
seed key import -n imported "word1 word2 ... word12"

# List all keys
seed key list

# Derive account ID without storing
seed key derive "word1 word2 ... word12"

Output Formats

The CLI supports multiple output formats:

JSON (default): Machine-readable structured data
YAML: Human-readable structured data (--yaml)
Pretty: Formatted output (--pretty)
Markdown: For documents (--md with get command)
Quiet: Minimal output (--quiet)

Server Configuration

By default, the CLI connects to https://hyper.media. You can:

Use a different server for a single command: -s, --server <url>
Set a default server: npx -y @seed-hypermedia/cli config --server <url>
View current configuration: npx -y @seed-hypermedia/cli config --show