ubie-oss/n8n-cli/.claude/skills/n8n-cli/SKILL.md
n8n-cli-operations
Build and operation support for n8n-cli. Guides remote status checks, imports, dry-runs, apply, and linter execution. Auto-build recommended for new sessions.
- Source repository stars
- 45
- Declared platforms
- 0
- Static risk flags
- 2
- Last source update
- 2026-08-25
- Source checked
- 2026-08-28
Decision brief
What it does: where it fits
A skill to assist with workflow management using n8n-cli (TypeScript/Bun).
Not for
- Authentication Error
- Network Error
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/ubie-oss/n8n-cli --skill ".claude/skills/n8n-cli"Inspect the Agent Skill "n8n-cli-operations" from https://github.com/ubie-oss/n8n-cli/blob/a0afecce09b575abef527324ed93192d0d5d26b0/.claude/skills/n8n-cli/SKILL.md at commit a0afecce09b575abef527324ed93192d0d5d26b0. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
Environment Setup
If .env exists at the project root, n8n-cli can be used without specifying parameters. Do not read the contents.
If .env exists at the project root, n8n-cli can be used without specifying parameters. Do not read the contents. - 02
1. Checking Remote Workflow Status
Check a specific workflow: bash
Check a specific workflow: bash - 03
Get workflow info by ID
Review the “Get workflow info by ID” section in the pinned source before continuing.
Review and apply the “Get workflow info by ID” source section. - 04
JSON format workflow
Review the “JSON format workflow” section in the pinned source before continuing.
Review and apply the “JSON format workflow” source section. - 05
YAML format workflow
./n8n-cli import --ids= --yaml
./n8n-cli import --ids= --yaml
Permission review
Static risk signals and limitations
Runs scripts
The documentation asks the agent to run terminal commands or scripts.
*For new sessions, always run the following first to ensure n8n-cli is up to date.**Runs scripts
The documentation asks the agent to run terminal commands or scripts.
bun install && make buildWrites files
The documentation asks the agent to create, modify, or delete local files.
Prompt the user to create a `.env` fileWrites files
The documentation asks the agent to create, modify, or delete local files.
# Save to file (useful for diff comparison)Evidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 92/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 45 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- ubie-oss/n8n-cli
- Skill path
- .claude/skills/n8n-cli/SKILL.md
- Commit
- a0afecce09b575abef527324ed93192d0d5d26b0
- License
- MIT
- Collected
- 2026-08-28
- Default branch
- main
View the original SKILL.md
n8n-cli Operations Skill
A skill to assist with workflow management using n8n-cli (TypeScript/Bun).
Session Initialization (Important)
For new sessions, always run the following first to ensure n8n-cli is up to date.
Priority 1: Build from Local Source (Development)
If the n8n-cli source repository is available via add-dir (e.g., /path/to/n8n-cli/Makefile exists), build from local source instead of using a release. This ensures the latest development changes are reflected immediately.
# If n8n-cli repo is visible via add-dir
N8N_CLI_REPO="/path/to/n8n-cli" # Replace with add-dir path
if [ -f "$N8N_CLI_REPO/Makefile" ]; then
echo "Building n8n-cli from local source: $N8N_CLI_REPO"
cd "$N8N_CLI_REPO" && bun install && make build
cp "$N8N_CLI_REPO/n8n-cli" /path/to/target/n8n-cli
cd /path/to/target
echo "Built from local source: $(./n8n-cli version 2>/dev/null | head -1)"
fi
Priority 2: Auto-Update from GitHub Releases (Preferred)
Pre-built binaries are available on GitHub Releases. This is faster than building from source and ensures you're running the official release.
# 1. Check the latest release version
LATEST_TAG=$(gh release view --repo ubie-oss/n8n-cli --json tagName -q '.tagName' 2>/dev/null)
# 2. Check current local version (if binary exists)
CURRENT_VERSION=""
if [ -x ./n8n-cli ]; then
CURRENT_VERSION=$(./n8n-cli version 2>/dev/null | head -1 | grep -oE 'v[0-9]+\.[0-9]+\.[0-9]+' || echo "")
fi
# 3. Compare and download if needed
if [ -z "$CURRENT_VERSION" ] || [ "$CURRENT_VERSION" != "$LATEST_TAG" ]; then
echo "Updating n8n-cli: ${CURRENT_VERSION:-none} -> ${LATEST_TAG}"
# Detect platform
ARCH=$(uname -m)
OS=$(uname -s | tr '[:upper:]' '[:lower:]')
case "$ARCH" in
arm64|aarch64) ASSET_NAME="n8n-cli-${OS}-arm64" ;;
x86_64) ASSET_NAME="n8n-cli-${OS}-x64" ;;
*) echo "Unsupported architecture: $ARCH"; exit 1 ;;
esac
# Download the binary
gh release download "$LATEST_TAG" --repo ubie-oss/n8n-cli --pattern "$ASSET_NAME" --output n8n-cli --clobber
chmod +x ./n8n-cli
echo "Updated to $(./n8n-cli version 2>/dev/null | head -1)"
else
echo "n8n-cli is up to date: $CURRENT_VERSION"
fi
Fallback: Build from Source
If gh CLI is not available or you need a development build:
# Install dependencies + build CLI
bun install && make build
Build verification:
./n8n-cli version
Environment Setup
Checking the .env File
If .env exists at the project root, n8n-cli can be used without specifying parameters. Do not read the contents.
# Only check for .env existence (do not read contents)
test -f .env && echo ".env exists - n8n-cli can be used without parameters"
If .env exists:
N8N_API_URLandN8N_API_KEYare already configured./n8n-clican be run without parameters
If .env does not exist:
- Prompt the user to create a
.envfile - Or use
--api-urland--api-keyflags
Creating .env (Only If Needed)
cp .env.example .env
# Ask the user to edit
Common Commands
1. Checking Remote Workflow Status
Check a specific workflow:
# Get workflow info by ID
./n8n-cli workflow get <workflow-id>
# Display in table format for readability
./n8n-cli -o table workflow get <workflow-id>
# Save to file (useful for diff comparison)
./n8n-cli workflow get <workflow-id> > /tmp/remote-workflow.json
List workflows:
# All workflows
./n8n-cli workflow list
# Table format
./n8n-cli -o table workflow list
# Active only
./n8n-cli workflow list --active
2. Import Workflows (Remote -> Local)
Check Local Format Before Importing
When pulling remote content to local, always follow these steps:
-
Check local file format
# Check if file exists and its format ls -la definitions/*<workflow-id>* definitions/**/*<workflow-id>* -
Use appropriate import options based on format
.jsonfiles: use--idsonly.yaml/.ymlfiles: use both--idsand--yaml
Import a specific workflow:
# JSON format workflow
./n8n-cli import --ids=<workflow-id>
# YAML format workflow
./n8n-cli import --ids=<workflow-id> --yaml
# To a custom directory
./n8n-cli import --ids=<workflow-id> -d ./my-dir
Import multiple workflows:
# Comma-separated list
./n8n-cli import --ids=abc123,def456,ghi789
Import all workflows:
./n8n-cli import
Additional options:
| Option | Description |
|---|---|
--include-archived | Include archived workflows |
--cleanup-orphans | Delete orphan files without matching IDs |
-t, --threshold <int> | Minimum lines for external file extraction (default: 0, configurable in CLAUDE.md) |
--dry-run | Preview only (no file changes) |
--tags <tags> | Filter by comma-separated tag names (AND condition) |
3. Dry-run (Change Preview)
Always verify with dry-run before applying:
# Preview changes for all workflows
./n8n-cli apply --dry-run
# Specific workflow only (by ID)
./n8n-cli apply --dry-run --ids=<workflow-id>
# Specific file only
./n8n-cli apply --dry-run -d definitions/path/to/workflow.json
Example output:
=== CREATE (1 workflow) ===
+ new-workflow.json (name: "New Automation")
=== UPDATE (2 workflows) ===
~ existing.json (id: R2cTI0LDzCJSnvNG)
- name: "Old Name" -> "New Name"
- nodes: 3 -> 5 nodes
=== SKIP (1 workflow) ===
= unchanged.json (no changes)
Summary (dry-run): 1 to create, 2 to update, 1 unchanged
4. Apply (Deploy)
AI assistants must always use --ids
# Apply specific workflow only (recommended)
./n8n-cli apply --ids=<workflow-id>
# Apply a specific file only
./n8n-cli apply -d definitions/path/to/workflow.json
# Force apply (overwrite remote changes)
./n8n-cli apply --ids=<workflow-id> --force
Prohibited: Applying to an entire directory
# These must NOT be run by AI
./n8n-cli apply -d definitions/
./n8n-cli apply
Additional options:
| Option | Description |
|---|---|
--from-git-changes <spec> | Apply only files changed in Git diff |
--yaml / --no-yaml | Enable/disable YAML processing |
--allow-duplicates | Skip the upstream duplicate-name check (default: on; use --force to push through warnings instead of disabling the check) |
--no-auto-tag | Disable automatic tagging (managed-as-code) |
-p, --project <id> | Specify target project ID |
Environment variables:
APPLY_FILTER_BY_TAGS- Filter by comma-separated tag names (AND condition)
Git diff mode:
# Apply only changes from a feature branch
./n8n-cli apply --from-git-changes origin/main..HEAD
# Preview
./n8n-cli apply --from-git-changes origin/main..HEAD --dry-run
# Changes from last 3 commits
./n8n-cli apply --from-git-changes HEAD~3..HEAD
3-way conflict detection: When --from-git-changes is specified, conflicts are detected by comparing Base (reference point), Local (current), and Remote (server) to identify only true conflicts.
5. Linter (Quality Checks)
The linter is integrated into the CLI. No separate build needed.
Note: Positional arguments are not supported. Always specify files/directories with -f or -d.
# Check a specific file (-f required)
./n8n-cli lint -f definitions/<filename>
# Check a specific directory
./n8n-cli lint -d ./definitions/example-project
# Check all workflows
./n8n-cli lint -d ./definitions
# JSON output (for CI/CD)
./n8n-cli lint -d ./definitions -o json
Options:
| Option | Description |
|---|---|
-d, --dir <path> | Definition files directory |
-f, --file <files...> | Specific file(s) |
-c, --config <path> | Config file path (.n8nlintrc.json) |
--disable-rule <name> | Disable specific rule(s) |
--list-rules | List all rules |
-o, --output <format> | Output format: text, json (default: text) |
--project <id> | Project ID context for local files |
Rules (11 rules):
| Rule | Severity | Description |
|---|---|---|
json-syntax | error | JSON syntax check |
required-fields | error | Required fields (name, nodes, connections) check |
connection-reference | error | Connection target node existence check |
webhook-id-required | error | Check that webhook and formTrigger nodes have webhookId field |
orphaned-node | warning | Orphaned node detection |
implicit-json-ref | warning | Implicit $json reference detection |
expression-mode-prefix | warning | Missing = prefix detection |
ai-agent-output-ref | warning | AI Agent output misreference detection |
node-params | warning | Node parameter schema validation |
node-ref-field-check | warning | Referenced field existence validation |
node-ref-cardinality | warning | .item/.first() usage validation |
Config file (.n8nlintrc.json):
{
"rules": {
"json-syntax": "error",
"required-fields": "error",
"orphaned-node": "warning",
"implicit-json-ref": "off"
},
"projects": {
"<n8n-project-id>": {
"rules": {
"banned-node": ["error", {
"nodes": [{ "type": "n8n-nodes-base.code" }]
}]
}
}
}
}
Top-level rules are global. Rules under a matching Project ID are an additional
layer and cannot disable global policy. Remote lint detects the owner from the
workflow; use --project for local files without ownership metadata.
Config file search order:
- Path specified with
--configflag .n8nlintrc.json/.n8nlintrcin the target directory.n8nlintrc.json/.n8nlintrcin the current directory
6. Formatter (Auto-format)
Note: fmt takes positional arguments for files (no -f option).
# Format a specific file (positional argument)
./n8n-cli fmt definitions/<filename>
# Multiple files
./n8n-cli fmt definitions/file1.yaml definitions/file2.yaml
# Format an entire directory
./n8n-cli fmt -d ./definitions
# Preview only
./n8n-cli fmt -d ./definitions --dry-run
7. Test (Execution)
Test workflows via test webhooks:
# Basic test
./n8n-cli test <workflow-id>
# With test data
./n8n-cli test <workflow-id> -d '{"email": "[email protected]"}'
# Wait for execution to complete (recommended)
./n8n-cli test <workflow-id> --wait-execution
# Auto-activate inactive workflows
./n8n-cli test <workflow-id> --activate
# Check webhook URL only
./n8n-cli test <workflow-id> --dry-run
# Check input parameters
./n8n-cli test <workflow-id> --show-inputs
Options:
| Option | Description |
|---|---|
-d, --data <json> | Test data (JSON string) |
--timeout <duration> | HTTP timeout (default: 30s) |
--wait-execution | Wait for execution to complete and show results |
--activate | Auto-activate inactive workflows |
--dry-run | Check webhook URL only |
--show-inputs | Display input parameters |
-o, --output <fmt> | Output format: json, table (default: json) |
Test from a local file:
# Specify a local JSON/YAML file
./n8n-cli test ./definitions/my-workflow.json
8. Execution (Logs & Errors)
List recent executions:
# List all recent executions
./n8n-cli execution list
# Table format for readability
./n8n-cli -o table execution list
# Filter by status
./n8n-cli execution list --status error
./n8n-cli execution list --status success
# Filter by workflow
./n8n-cli execution list --workflow <workflow-id>
# Limit results
./n8n-cli execution list --limit 5
# Combined filters
./n8n-cli execution list --workflow <workflow-id> --status error --limit 10
Get execution details:
# Get execution by ID
./n8n-cli execution get <execution-id>
# Table format with error details
./n8n-cli -o table execution get <execution-id>
# Include node execution summary
./n8n-cli -o table execution get <execution-id> --show-data
Options for execution list:
| Option | Description |
|---|---|
-w, --workflow <id> | Filter by workflow ID |
-s, --status <status> | Filter by status (success, error, running, waiting) |
-l, --limit <n> | Maximum number of executions (default: 20) |
Options for execution get:
| Option | Description |
|---|---|
--show-data | Include node execution summary in output |
Error information displayed:
| Field | Description |
|---|---|
| Error Node | The node where the error occurred |
| Error Message | The error message |
| Error Details | Additional error description (if available) |
| Last Node | The last executed node |
Typical debugging workflow:
# 1. List recent errors
./n8n-cli -o table execution list --status error --limit 5
# 2. Get details of a specific failed execution
./n8n-cli -o table execution get <execution-id> --show-data
# 3. Review error details and fix the workflow
9. Convert (Format Conversion)
Convert workflow files between JSON and YAML formats (local-only, no API needed):
# Convert all JSON workflows to YAML
./n8n-cli convert -d ./definitions --format yaml
# Convert specific workflows by ID
./n8n-cli convert -d ./definitions --format json --ids <workflow-id>
# Preview without writing
./n8n-cli convert -d ./definitions --format yaml --dry-run
# Keep original files
./n8n-cli convert -d ./definitions --format yaml --keep
# Convert a single file
./n8n-cli convert --format yaml definitions/<filename>.json
Options:
| Option | Description |
|---|---|
--format <format> | Target format: json, yaml (required) |
-d, --directory <dir> | Directory to scan for workflow files |
--ids <ids> | Comma-separated workflow IDs to convert |
--tags <tags> | Filter by tags (comma-separated, AND condition) |
-t, --threshold <n> | Minimum lines for code externalization (JSON→YAML) |
--dry-run | Preview only |
--keep | Keep original files |
Behavior:
- JSON→YAML: generates YAML +
_subfiles/with externalized code - YAML→JSON: resolves
!includerefs and removes_subfiles/ - Files already in target format are skipped
- Original files removed after conversion unless
--keep
10. Data Tables
Manage data tables and rows:
# List all data tables
./n8n-cli data-tables list
./n8n-cli -o table data-tables list
# Get a data table by ID (includes column definitions)
./n8n-cli data-tables get <data-table-id>
# Create a data table
./n8n-cli data-tables create --name "My Table" --columns '[{"name":"col1","type":"string"},{"name":"col2","type":"number"}]'
# Update a data table name
./n8n-cli data-tables update <data-table-id> --name "New Name"
# Delete data tables
./n8n-cli data-tables delete <data-table-id> --force
Row operations:
# List rows
./n8n-cli data-tables rows list <data-table-id>
./n8n-cli data-tables rows list <data-table-id> --limit 10 --search "keyword"
# Insert rows
./n8n-cli data-tables rows insert <data-table-id> --data '[{"col1":"hello","col2":42}]'
./n8n-cli data-tables rows insert <data-table-id> --data '[{"col1":"value"}]' --return-type all
# Update rows matching a filter
./n8n-cli data-tables rows update <data-table-id> \
--filter '{"type":"and","filters":[{"columnName":"col1","condition":"eq","value":"hello"}]}' \
--data '{"col2":99}' --dry-run
# Upsert rows
./n8n-cli data-tables rows upsert <data-table-id> \
--filter '{"type":"and","filters":[{"columnName":"col1","condition":"eq","value":"hello"}]}' \
--data '{"col1":"hello","col2":100}'
# Delete rows matching a filter
./n8n-cli data-tables rows delete <data-table-id> \
--filter '{"type":"and","filters":[{"columnName":"col1","condition":"eq","value":"hello"}]}' \
--force
Options:
| Subcommand | Key Options |
|---|---|
list | --limit, --filter <json>, --sort-by <field:dir> |
rows list | --limit, --filter <json>, --sort-by, --search <text> |
rows insert | -d, --data <json> (required), --return-type count|id|all |
rows update | --filter <json> (required), -d, --data <json> (required), --return-data, --dry-run |
rows upsert | --filter <json> (required), -d, --data <json> (required), --return-data, --dry-run |
rows delete | --filter <json> (required), --return-data, --dry-run, --force |
Supported column types: string, number, boolean, date, json.
11. Trace (Data Flow Analysis)
Analyze data flow and cardinality through a workflow:
# Trace a workflow file
./n8n-cli trace -f definitions/<filename>
# JSON output
./n8n-cli trace -f definitions/<filename> -o json
Output columns:
| Column | Description |
|---|---|
| Node | Node name |
| Type | n8n node type |
| Cardinality | Output cardinality: 1:1, 1:N, N:1, pass-through, variable, unknown |
| Items | Estimated output item count: 1, N, ?, loop, etc. |
| Inputs | Upstream nodes |
| Outputs | Downstream nodes |
Cardinality meanings:
| Cardinality | Meaning |
|---|---|
1:1 | One input item produces one output item |
1:N | One input item produces multiple output items |
N:1 | Multiple input items produce one output item (e.g. Aggregate) |
pass-through | Items pass through unchanged (e.g. Filter, If, Set) |
variable | Output count depends on runtime behavior (e.g. Code, HTTP Request) |
unknown | Node type has no cardinality definition |
Interpreting ? in estimated items:
When a node shows ? for estimated items, it means cardinality could not be statically determined. AI assistants should:
- Check the node's
operationparameter — many nodes (Notion, GoogleSheets, BigQuery) have operation-dependent cardinality that the trace already resolves - Check
limitparameters or SQLLIMITclauses that may constrain output - For
Codenodes, read the code to determine if it produces 1 or N items - For
HTTP Request, check if the response is an array or single object - Do not assume
?means "many" — it simply means "unknown at static analysis time"
12. Credential (Credential Management)
Check available credentials:
# List all credentials
./n8n-cli credential list
# Get a specific credential's details
./n8n-cli credential get <id>
# Check credential type schema
./n8n-cli credential schema <typeName>
When designing workflows:
- If a node uses an external service, verify the corresponding credential exists
- If a credential is missing, ask the user to create it in the n8n UI
13. Node Schema (Node Schema Reference)
Check node parameter definitions:
# List all nodes
./n8n-cli node-schema list
# JSON output
./n8n-cli node-schema list --output json
# Dump a specific node's schema
./n8n-cli node-schema dump --type n8n-nodes-base.slack
# Dump all nodes to a directory (individual JSON + _index.json)
./n8n-cli node-schema dump -o ./schemas
Prerequisites: n8n-nodes-base and @n8n/n8n-nodes-langchain must be installed in node_modules.
Options (list):
| Option | Description |
|---|---|
--output json | JSON output |
--group <group> | Filter by group (trigger, transform, etc.) |
Options (dump):
| Option | Description |
|---|---|
--type <nodeType> | Specific node type schema (e.g., n8n-nodes-base.slack) |
-o, --output-dir <dir> | Dump all nodes as individual files to directory |
Typical Workflow Operations
Editing Workflows
# 1. Update CLI (auto-update from GitHub Releases, fallback: bun install && make build)
LATEST_TAG=$(gh release view --repo ubie-oss/n8n-cli --json tagName -q '.tagName' 2>/dev/null)
CURRENT_VERSION=$(./n8n-cli version 2>/dev/null | head -1 | grep -oE 'v[0-9]+\.[0-9]+\.[0-9]+' || echo "")
if [ -z "$CURRENT_VERSION" ] || [ "$CURRENT_VERSION" != "$LATEST_TAG" ]; then
ARCH=$(uname -m); OS=$(uname -s | tr '[:upper:]' '[:lower:]')
case "$ARCH" in arm64|aarch64) A="n8n-cli-${OS}-arm64";; x86_64) A="n8n-cli-${OS}-x64";; esac
gh release download "$LATEST_TAG" --repo ubie-oss/n8n-cli --pattern "$A" --output n8n-cli --clobber && chmod +x ./n8n-cli
fi
# 2. Check .env
test -f .env && echo "OK"
# 3. Edit local file (using n8n-workflow skill)
# 4. Lint check
./n8n-cli lint -f definitions/<filename>
# 5. Dry-run to verify
./n8n-cli apply --dry-run --ids=<workflow-id>
# 6. Apply
./n8n-cli apply --ids=<workflow-id>
# 7. Post-apply dry-run to confirm results
./n8n-cli apply --dry-run --ids=<workflow-id>
# 8. Test (optional)
./n8n-cli test <workflow-id> --wait-execution
Pulling Remote Changes
# 1. Check local file format (important!)
ls -la definitions/*<workflow-id>* definitions/**/*<workflow-id>*
# 2. Check remote state
./n8n-cli workflow get <workflow-id> > /tmp/remote.json
# 3. Compare with local (if needed)
diff definitions/path/to/workflow.json /tmp/remote.json
# 4. Import preserving format
# For JSON format:
./n8n-cli import --ids=<workflow-id>
# For YAML format:
./n8n-cli import --ids=<workflow-id> --yaml
Exit Codes
| Code | Meaning |
|---|---|
| 0 | Success |
| 1 | Error (invalid JSON, API error, network failure) |
| 2 | Conflict detected (dry-run only, when remote is newer) or rule violation (lint error) |
Troubleshooting
Authentication Error
Error: authentication failed
-> Check N8N_API_KEY in .env
Network Error
Error: network error
-> Check N8N_API_URL in .env
Conflict Error
Error: conflict: remote workflow has been modified since your local file
-> Check remote changes and either overwrite with --force or import and re-edit
Guidelines for AI Assistants
At Session Start
-
Auto-update n8n-cli from GitHub Releases
# Check latest release and download if newer LATEST_TAG=$(gh release view --repo ubie-oss/n8n-cli --json tagName -q '.tagName' 2>/dev/null) CURRENT_VERSION="" if [ -x ./n8n-cli ]; then CURRENT_VERSION=$(./n8n-cli version 2>/dev/null | head -1 | grep -oE 'v[0-9]+\.[0-9]+\.[0-9]+' || echo "") fi if [ -z "$CURRENT_VERSION" ] || [ "$CURRENT_VERSION" != "$LATEST_TAG" ]; then ARCH=$(uname -m); OS=$(uname -s | tr '[:upper:]' '[:lower:]') case "$ARCH" in arm64|aarch64) ASSET="n8n-cli-${OS}-arm64" ;; x86_64) ASSET="n8n-cli-${OS}-x64" ;; esac gh release download "$LATEST_TAG" --repo ubie-oss/n8n-cli --pattern "$ASSET" --output n8n-cli --clobber && chmod +x ./n8n-cli fi ./n8n-cli versionghCLI が使えない場合のフォールバック:bun install && make build
-
Check .env existence (do not read contents)
test -f .env && echo ".env exists"
After Editing Workflows (Development Feedback Loop)
- Lint check -
./n8n-cli lint -f definitions/<filename> - Dry-run to verify changes -
./n8n-cli apply --dry-run --ids=<workflow-id> - Proactively apply for testing (always specify a single workflow)
- Unless the user explicitly declines, apply even during development
- After apply, share the n8n UI URL
- If issues are found, repeat the fix -> apply cycle
- Post-apply dry-run - Confirm no conflicts
When Importing from Remote
Important: Preserve the local format
-
Check local file format
ls -la definitions/*<workflow-id>* definitions/**/*<workflow-id>* -
Determine format
.json-> JSON format.yaml/.yml-> YAML format
-
Import with format-appropriate options
# JSON format ./n8n-cli import --ids=<workflow-id> # YAML format ./n8n-cli import --ids=<workflow-id> --yaml
Prohibited Actions
- Reading the contents of
.envfiles - Applying to an entire directory (always use
--ids) - Using
--forcecarelessly - Importing without checking the local format first
Frequently asked questions
What to verify before installation and use
What does the n8n-cli-operations source document cover?
A skill to assist with workflow management using n8n-cli (TypeScript/Bun).
How do I install n8n-cli-operations?
The source record exposes this install command: npx skills add https://github.com/ubie-oss/n8n-cli --skill ".claude/skills/n8n-cli". Inspect the command and pinned source before running it.
Which permission-related actions were detected?
Static rules flagged exec-script, write-files in the source; the page lists the matching lines and excerpts.
Alternatives
Compare before choosing
Postpartum-genushyacinthus29/dotnet-skills
dotnet-worker-services
Build long-running .NET background services with `BackgroundService`, Generic Host, graceful shutdown, configuration, logging, and deployment patterns suited to workers and daemons.
vasilyu1983/AI-Agents-public
qa-testing-ios
Guides iOS testing with XCTest, XCUITest, Swift Testing, simctl, and xcresult. Use when choosing destinations, controlling flakes, or parsing test artifacts for native apps.
garrytan/gbrain
brain-pdf
Generate a publication-quality PDF from any brain page via the gstack make-pdf binary. Strips YAML frontmatter, sanitizes emoji, applies running headers and page numbers. Brain page is always the source of truth; PDF is a rendering.
NVIDIA/skills
rtvi-cv-customize-model
How to swap the DeepStream CV detection model in the VSS Alerts Blueprint verification (2d_cv) mode - covers ONNX export, custom bbox parsers, compose mount gotchas, nvinfer config, runtime TRT engine build, deployment, and a segmentation-capable model addendum handoff.