Use Orchata CLI commands to manage knowledge bases from the terminal. For shell/terminal operations only.
Install
npx skillscat add orchata-ai/orchata-skills/orchata-cli Install via the SkillsCat registry.
Orchata CLI Commands
Use this skill when you need to run orchata commands in a terminal/shell environment.
Use when:
- You need to run shell commands in a terminal
- You need to perform batch file uploads
- You're working with file system operations
- You need to script or automate Orchata operations
- You want JSON output for parsing
Don't use when:
- You have MCP tools available (use
orchata-mcpskill instead) - You need tree-based document navigation (MCP only feature)
- You need programmatic function calls (use
orchata-mcpskill instead)
What is Orchata?
Orchata is a knowledge management platform that:
- Organizes documents into Spaces - Logical containers for related content
- Provides semantic search - Find relevant content using natural language queries
- Provides CLI - Terminal interface for document management and queries
Installation & Setup
Installation
# Install globally via npm
npm install -g @orchata-ai/cli
# Or via bun
bun add -g @orchata-ai/cli
# Verify installation
orchata --versionInitial Configuration
# 1. Configure CLI (run once after install)
orchata init
# 2. Authenticate with your API key
orchata login
# 3. Verify connection
orchata spaces list
# 4. (Optional) Set a default space to avoid --space on every command
orchata workspace setEnvironment Variables
# Optional: Override settings via environment variables
export ORCHATA_API_KEY="oai_..."
export ORCHATA_API_BASE="https://api.orchata.ai"
export ORCHATA_PROFILE="production"CLI Commands Reference
Workspace Context
Set a default space so you don't need --space on every command.
Set default space (interactive picker)
orchata workspace setShows an interactive list of your spaces to choose from.
Set default space by ID
orchata workspace set space_abc123Show current default
orchata workspace showClear default
orchata workspace clearOnce a workspace is set, all commands that normally require --space will use the default automatically. You can still override with --space <id> on any command.
Space Management
List all spaces
orchata spaces listCreate a space
orchata spaces create --name "Product Docs" --description "Technical documentation" --icon bookOptions:
--name <name>- Space name (required)--description <text>- Space description (optional)--icon <icon>- Icon:folder,book,file-text,database,package,archive,briefcase,inbox,layers,box
Get a specific space
orchata spaces get space_abc123Update a space
orchata spaces update space_abc123 --name "Updated Name" --description "New description"Delete a space (soft delete/archive)
orchata spaces delete space_abc123Document Management
List documents in a space
orchata documents list --space space_abc123
# Or with a default workspace set:
orchata documents listUpload a single file
orchata documents upload ./report.pdf --space space_abc123
orchata documents upload ./guide.mdUpload multiple files with glob patterns
# Upload all markdown files in a directory
orchata documents upload docs/*.md --space space_abc123
# Upload with metadata applied to all files
orchata documents upload notes/*.md --metadata '{"tag":"drafts"}'Files are batched automatically (up to 100 per request). A summary is shown:
Uploaded 12 file(s)Upload inline content
orchata documents upload --content "# Title\n\nContent here..." --filename my-doc.mdBatch upload from JSON file
orchata documents batch --file ./docs.json --space space_abc123Upsert (create or update by filename)
orchata documents upsert --filename "brief.md" --content "# Updated content..."If the filename already exists in the space, it updates the document. If new, it creates it.
Get a document
orchata documents get doc_xyz789 --space space_abc123Get document content (processed text)
orchata documents content doc_xyz789 --space space_abc123Append to a document
orchata documents append doc_xyz789 --content "Additional content..."Get by filename
orchata documents get-by-filename --filename "guide.md"Querying
Query with compact output
orchata query --compact "How do I authenticate?"Output:
3 result(s):
api-guide.md - Authentication
"To authenticate API requests, include your API key in the header..."
handbook.pdf - Security Overview
"All API calls require authentication via Bearer token..."Query with full JSON output
orchata query "How do I authenticate?" --space space_abc123Query with more results
orchata query "installation guide" --top-k 15Options:
--space <id>- Space ID to query (uses workspace default if set, or all spaces)--compact- Human-readable output instead of JSON--top-k <n>- Maximum number of results (default: 10)--threshold <n>- Similarity threshold 0-1 (default: 0)--group-by-space- Group results by space--metadata <json>- Metadata filter
Smart query (discover relevant spaces)
orchata query smart "what is orchata"This helps you find which spaces are relevant when you don't know where to look.
Global Options
These options work with any command:
orchata spaces list --profile production
orchata query "test" --space space_123 --api-key oai_custom_key --json
orchata documents delete doc_123 --yesAvailable global options:
--profile <name>- Use a named profile--api-base <url>- Override API base URL--app-base <url>- Override app base URL--api-key <key>- Override API key for this run--json- Output raw JSON (useful for scripting)-y, --yes- Auto-confirm all prompts (non-interactive mode, ideal for agents/scripts)
Common Patterns
Pattern 1: First-time setup with workspace context
# 1. Authenticate
orchata login
# 2. Create a space
orchata spaces create --name "Docs" --description "Product documentation"
# Returns: space_abc123
# 3. Set it as your default workspace
orchata workspace set space_abc123
# 4. Now all commands use this space automatically
orchata documents upload ./handbook.pdf
orchata documents upload ./api-guide.md
orchata query --compact "authentication flow"Pattern 2: Bulk upload with glob patterns
# Upload all markdown files
orchata documents upload ./documentation/*.md
# Upload with metadata
orchata documents upload ./drafts/*.md --metadata '{"source":"drafts"}'Pattern 3: Discover and search
# 1. Find relevant spaces
orchata query smart "billing questions"
# Returns: space_billing (most relevant)
# 2. Query that space
orchata query --compact "payment methods" --space space_billingPattern 4: Get raw JSON for scripting
# Get JSON output for parsing
orchata spaces list --json | jq '.spaces[] | .id'
orchata query "test" --space space_123 --json | jq '.results[0]'Pattern 5: Non-interactive mode for agents/scripts
# Auto-confirm all prompts with --yes
orchata workspace set --yes # Picks first space
orchata documents delete doc_123 --yes # No confirmation prompt
# Combine with --json for fully deterministic automation
orchata query "search term" --json --yesPattern 6: Profile management for multiple environments
# Login to different environments
orchata login --profile development
orchata login --profile production
# Use specific profile
orchata query "test" --space space_123 --profile productionDocument Processing
Documents are processed asynchronously:
- Upload returns immediately with
status="PROCESSING" - Background job generates embeddings and indexes the document (typically 1-3 seconds)
- Status changes to
"COMPLETED"when ready - Document becomes searchable
To check completion status:
# List all documents with their status
orchata documents list
# Check specific document
orchata documents get doc_xyz789Supported file formats:
- PDF (text-based and scanned with OCR)
- Word documents (.docx)
- Excel spreadsheets (.xlsx)
- PowerPoint presentations (.pptx)
- Markdown files (.md)
- Plain text files (.txt)
- Images (PNG, JPG, etc.)
Best Practices
DO
- Set a default workspace -
orchata workspace seteliminates--spaceon every command - Use glob upload for multiple files -
orchata documents upload docs/*.md - Use
--compactfor quick lookups - Human-readable query results - Use
--jsonflag for scripting - Easy to parse programmatically - Use
--yesfor automation - No interactive prompts in scripts/agents - Use
upsertfor iterative docs - Avoids duplicates when re-uploading - Wait 1-3 seconds after upload - Give documents time to process
- Check document status before querying - Only COMPLETED documents are searchable
DON'T
- Don't query immediately after upload - Wait for processing to complete
- Don't use very short queries - More context = better results
- Don't forget to authenticate - Run
orchata loginfirst - Don't memorize space IDs - Use
orchata workspace setfor interactive selection
Troubleshooting
"Command not found: orchata"
Solution:
npm install -g @orchata-ai/cli
orchata --version"Authentication required"
Solution:
orchata login
# Or set environment variable:
export ORCHATA_API_KEY="oai_...""No space specified"
Solution:
# Set a default workspace
orchata workspace set
# Or pass explicitly
orchata documents list --space space_abc123"Document still processing"
Solution:
Wait 1-3 seconds after upload for processing to complete:
orchata documents list
# Check status fieldUpload fails
Common causes:
- File is too large (check file size limits)
- Invalid file format (check supported formats above)
- Space is archived (use a different space)
- Network issues (check connection)
Configuration Files
The CLI stores configuration in ~/.orchata/config.json:
{
"currentProfile": "cloud",
"profiles": {
"cloud": {
"mode": "cloud",
"apiBase": "https://api.orchata.ai/api",
"appBase": "https://app.orchata.ai",
"apiKey": "oai_...",
"defaultSpace": "space_abc123"
},
"local": {
"mode": "local",
"apiBase": "http://localhost:4747/api",
"appBase": "http://localhost:4747"
}
}
}You can also manage configuration with:
orchata configure --api-base https://api.orchata.ai
orchata configure --profile production --set-defaultQuick Reference
| Task | Command |
|---|---|
| Set default space | orchata workspace set |
| Show default space | orchata workspace show |
| List spaces | orchata spaces list |
| Create space | orchata spaces create --name "Docs" |
| List documents | orchata documents list |
| Upload file | orchata documents upload ./file.pdf |
| Upload glob | orchata documents upload docs/*.md |
| Upsert document | orchata documents upsert --filename "x.md" --content "..." |
| Search (compact) | orchata query --compact "question" |
| Search (JSON) | orchata query "question" --json |
| Discover spaces | orchata query smart "question" |
| Non-interactive | orchata <command> --yes |
Script Examples
Bash: Upload and wait for processing
#!/bin/bash
SPACE_ID="space_abc123"
FILE="./document.pdf"
# Upload
echo "Uploading ${FILE}..."
RESULT=$(orchata documents upload "${FILE}" --space "${SPACE_ID}" --json)
DOC_ID=$(echo $RESULT | jq -r '.document.id')
echo "Document ID: ${DOC_ID}"
echo "Waiting for processing..."
# Wait for completion
while true; do
STATUS=$(orchata documents get "${DOC_ID}" --space "${SPACE_ID}" --json | jq -r '.document.status')
if [ "$STATUS" = "COMPLETED" ]; then
echo "Processing complete!"
break
elif [ "$STATUS" = "FAILED" ]; then
echo "Processing failed!"
exit 1
fi
echo "Status: ${STATUS}..."
sleep 2
done
# Query
orchata query --compact "summary of document" --space "${SPACE_ID}"Bash: Agent-friendly bulk ingest
#!/bin/bash
# Non-interactive: uses --yes and --json throughout
SPACE_ID="space_abc123"
# Upload all markdown files
orchata documents upload ./docs/*.md --space "${SPACE_ID}" --json --yes
# Wait and verify
sleep 3
orchata documents list --space "${SPACE_ID}" --json | jq '.documents[] | "\(.filename): \(.status)"'Differences from MCP Tools
CLI does NOT support:
- Tree-based document navigation (
get_document_tree,get_tree_node)
CLI excels at:
- Glob-based batch file uploads (
orchata documents upload docs/*.md) - Compact human-readable query output (
--compact) - Workspace context persistence (
orchata workspace set) - Non-interactive mode for agents (
--yes) - Scripting and automation (
--json) - Profile management for multiple environments
- File system integration
For tree-based navigation and programmatic access, use the orchata-mcp skill instead.