@path58/p58-n8n 0.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/AGENT_INSTALL.md +223 -0
- package/CHANGELOG.md +88 -0
- package/LICENSE +21 -0
- package/README.md +281 -0
- package/dist/mcp/server.bundle.cjs +44423 -0
- package/package.json +291 -0
package/AGENT_INSTALL.md
ADDED
|
@@ -0,0 +1,223 @@
|
|
|
1
|
+
# P58-n8n — Drop & Install
|
|
2
|
+
|
|
3
|
+
**For users:** Drop this file into any AI chat and say: **"Install this MCP server for me"**
|
|
4
|
+
|
|
5
|
+
**For AI agents:** Follow the instructions in the AI AGENT section below.
|
|
6
|
+
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
## What you'll get
|
|
10
|
+
|
|
11
|
+
P58-n8n gives your AI assistant deep n8n workflow knowledge — 1,982 nodes, 40,179 operations, 26 auto-fix rules.
|
|
12
|
+
|
|
13
|
+
| Tool | What it does |
|
|
14
|
+
|------|-------------|
|
|
15
|
+
| `validate_workflow` | Run L1-L6 validation on workflow JSON |
|
|
16
|
+
| `get_operation_schema` | Get exact parameter requirements for any operation |
|
|
17
|
+
| `check_parameter` | Validate a parameter value with typo detection |
|
|
18
|
+
| `suggest_fix` | Get fix suggestions for validation issues |
|
|
19
|
+
| `list_nodes` | Browse and search 1,982 n8n nodes |
|
|
20
|
+
| `list_operations` | List all operations for a node type |
|
|
21
|
+
| `get_node_info` | Get comprehensive node metadata |
|
|
22
|
+
| `find_similar_pattern` | Search 9,000+ validated workflows |
|
|
23
|
+
|
|
24
|
+
### Supported clients
|
|
25
|
+
|
|
26
|
+
| Client | Support | Transport |
|
|
27
|
+
|--------|---------|-----------|
|
|
28
|
+
| Claude Desktop (Chat + Cowork) | Primary | stdio |
|
|
29
|
+
| Claude Code (CLI) | Primary | stdio |
|
|
30
|
+
| Gemini CLI | Secondary | stdio |
|
|
31
|
+
| ChatGPT Desktop | Not supported | Requires HTTP/SSE |
|
|
32
|
+
|
|
33
|
+
---
|
|
34
|
+
|
|
35
|
+
## AI AGENT: Follow These Instructions
|
|
36
|
+
|
|
37
|
+
### STEP 1 — Detect Environment
|
|
38
|
+
|
|
39
|
+
Determine your capabilities and the user's platform.
|
|
40
|
+
|
|
41
|
+
**Can you execute commands?**
|
|
42
|
+
```
|
|
43
|
+
Can I run shell commands?
|
|
44
|
+
├─ YES (Claude Code, Cowork mode, Gemini CLI) → Go to STEP 2a (Auto-Install)
|
|
45
|
+
└─ NO (Claude Desktop Chat-only) → Go to STEP 2b (Guided Install)
|
|
46
|
+
```
|
|
47
|
+
|
|
48
|
+
**Detect platform:**
|
|
49
|
+
```
|
|
50
|
+
What OS is the user on?
|
|
51
|
+
├─ macOS → Config paths use ~/Library/Application Support/
|
|
52
|
+
├─ Windows → Config paths use %APPDATA%\
|
|
53
|
+
└─ Linux → Config paths use ~/.config/
|
|
54
|
+
```
|
|
55
|
+
|
|
56
|
+
---
|
|
57
|
+
|
|
58
|
+
### STEP 2a — Auto-Install (can run commands)
|
|
59
|
+
|
|
60
|
+
**Prerequisites — verify first:**
|
|
61
|
+
|
|
62
|
+
```bash
|
|
63
|
+
node --version
|
|
64
|
+
# Must be v18.0.0 or higher. If not installed, tell the user to install from https://nodejs.org/
|
|
65
|
+
```
|
|
66
|
+
|
|
67
|
+
**Then install based on which client the user wants:**
|
|
68
|
+
|
|
69
|
+
#### Claude Code
|
|
70
|
+
|
|
71
|
+
Run this single command:
|
|
72
|
+
|
|
73
|
+
```bash
|
|
74
|
+
claude mcp add p58-n8n -- npx -y @path58/p58-n8n
|
|
75
|
+
```
|
|
76
|
+
|
|
77
|
+
Done. Skip to STEP 3.
|
|
78
|
+
|
|
79
|
+
#### Claude Desktop (Cowork mode — can write files)
|
|
80
|
+
|
|
81
|
+
Read the existing config file first, then MERGE the p58-n8n entry. NEVER overwrite the entire file.
|
|
82
|
+
|
|
83
|
+
**Config file paths:**
|
|
84
|
+
- macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
|
|
85
|
+
- Windows: `%APPDATA%\Claude\claude_desktop_config.json`
|
|
86
|
+
- Linux: `~/.config/Claude/claude_desktop_config.json`
|
|
87
|
+
|
|
88
|
+
**If file exists:** Read it, parse the JSON, add this entry inside `mcpServers`:
|
|
89
|
+
|
|
90
|
+
```json
|
|
91
|
+
"p58-n8n": {
|
|
92
|
+
"command": "npx",
|
|
93
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
94
|
+
}
|
|
95
|
+
```
|
|
96
|
+
|
|
97
|
+
**If file doesn't exist:** Create it with:
|
|
98
|
+
|
|
99
|
+
```json
|
|
100
|
+
{
|
|
101
|
+
"mcpServers": {
|
|
102
|
+
"p58-n8n": {
|
|
103
|
+
"command": "npx",
|
|
104
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
105
|
+
}
|
|
106
|
+
}
|
|
107
|
+
}
|
|
108
|
+
```
|
|
109
|
+
|
|
110
|
+
Tell the user to restart Claude Desktop.
|
|
111
|
+
|
|
112
|
+
#### Gemini CLI
|
|
113
|
+
|
|
114
|
+
There is NO `gemini mcp add` command. Edit the config file directly.
|
|
115
|
+
|
|
116
|
+
**Config path:** `~/.gemini/settings.json` (user-level) or `.gemini/settings.json` (project-level)
|
|
117
|
+
|
|
118
|
+
**If file exists:** Read it, parse JSON, add the `p58-n8n` entry inside `mcpServers`.
|
|
119
|
+
|
|
120
|
+
**If file doesn't exist:** Create it with:
|
|
121
|
+
|
|
122
|
+
```json
|
|
123
|
+
{
|
|
124
|
+
"mcpServers": {
|
|
125
|
+
"p58-n8n": {
|
|
126
|
+
"command": "npx",
|
|
127
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
128
|
+
}
|
|
129
|
+
}
|
|
130
|
+
}
|
|
131
|
+
```
|
|
132
|
+
|
|
133
|
+
---
|
|
134
|
+
|
|
135
|
+
### STEP 2b — Guided Install (chat-only)
|
|
136
|
+
|
|
137
|
+
Ask the user which client they want to set up, then give them copy-paste instructions.
|
|
138
|
+
|
|
139
|
+
**Ask:** "Which AI client do you want to use P58-n8n with?"
|
|
140
|
+
|
|
141
|
+
#### For Claude Desktop
|
|
142
|
+
|
|
143
|
+
Tell the user:
|
|
144
|
+
|
|
145
|
+
1. Open your Claude Desktop config file:
|
|
146
|
+
- **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
|
|
147
|
+
- **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
|
|
148
|
+
- **Linux:** `~/.config/Claude/claude_desktop_config.json`
|
|
149
|
+
|
|
150
|
+
2. Add this inside the `mcpServers` object (create the file if it doesn't exist):
|
|
151
|
+
|
|
152
|
+
```json
|
|
153
|
+
{
|
|
154
|
+
"mcpServers": {
|
|
155
|
+
"p58-n8n": {
|
|
156
|
+
"command": "npx",
|
|
157
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
158
|
+
}
|
|
159
|
+
}
|
|
160
|
+
}
|
|
161
|
+
```
|
|
162
|
+
|
|
163
|
+
3. Restart Claude Desktop.
|
|
164
|
+
|
|
165
|
+
#### For Claude Code
|
|
166
|
+
|
|
167
|
+
Tell the user to run:
|
|
168
|
+
|
|
169
|
+
```bash
|
|
170
|
+
claude mcp add p58-n8n -- npx -y @path58/p58-n8n
|
|
171
|
+
```
|
|
172
|
+
|
|
173
|
+
#### For Gemini CLI
|
|
174
|
+
|
|
175
|
+
Tell the user:
|
|
176
|
+
|
|
177
|
+
1. Open `~/.gemini/settings.json` (create if needed)
|
|
178
|
+
2. Add:
|
|
179
|
+
|
|
180
|
+
```json
|
|
181
|
+
{
|
|
182
|
+
"mcpServers": {
|
|
183
|
+
"p58-n8n": {
|
|
184
|
+
"command": "npx",
|
|
185
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
186
|
+
}
|
|
187
|
+
}
|
|
188
|
+
}
|
|
189
|
+
```
|
|
190
|
+
|
|
191
|
+
#### For ChatGPT Desktop
|
|
192
|
+
|
|
193
|
+
Tell the user: **ChatGPT Desktop is not supported.** It requires HTTP/SSE transport for MCP servers, but P58-n8n uses stdio transport. This may be supported in a future version.
|
|
194
|
+
|
|
195
|
+
---
|
|
196
|
+
|
|
197
|
+
### STEP 3 — Verify Installation
|
|
198
|
+
|
|
199
|
+
After setup, test with this prompt:
|
|
200
|
+
|
|
201
|
+
> "List available n8n nodes for email"
|
|
202
|
+
|
|
203
|
+
**Expected result:** The AI calls `list_nodes` and returns nodes like `n8n-nodes-base.gmail`, `n8n-nodes-base.emailSend`, etc.
|
|
204
|
+
|
|
205
|
+
**If it doesn't work:**
|
|
206
|
+
|
|
207
|
+
```
|
|
208
|
+
Tools not appearing?
|
|
209
|
+
├─ Config file in wrong location → Check OS-specific paths above
|
|
210
|
+
├─ Invalid JSON syntax → Validate with: cat <path> | python3 -m json.tool
|
|
211
|
+
├─ Client not restarted → Close and reopen the AI client
|
|
212
|
+
├─ npx not found → Install Node.js 18+ from https://nodejs.org/
|
|
213
|
+
└─ Permission error → Try: sudo npm install -g @path58/p58-n8n
|
|
214
|
+
|
|
215
|
+
Server starts but tools don't work?
|
|
216
|
+
├─ Database not accessible → Server needs network access to Supabase
|
|
217
|
+
├─ Timeout errors → Check internet connection
|
|
218
|
+
└─ "Module not found" → Run: npm install -g @path58/p58-n8n (reinstall)
|
|
219
|
+
```
|
|
220
|
+
|
|
221
|
+
---
|
|
222
|
+
|
|
223
|
+
**Package:** `@path58/p58-n8n` | **License:** MIT | **Built by [Path58](https://path58.com)**
|
package/CHANGELOG.md
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
1
|
+
# Changelog
|
|
2
|
+
|
|
3
|
+
All notable changes to this project will be documented in this file.
|
|
4
|
+
|
|
5
|
+
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
|
|
6
|
+
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
|
|
7
|
+
|
|
8
|
+
## [0.2.0] - 2026-03-08
|
|
9
|
+
|
|
10
|
+
### Added
|
|
11
|
+
|
|
12
|
+
- **26 new MCP tools** — total now 34 across 7 tiers (up from 8 in v0.1.0)
|
|
13
|
+
- **Tier 2 — Server CRUD (7 tools):** `get_workflow`, `list_workflows`, `create_workflow`, `update_workflow`, `delete_workflow`, `execute_workflow`, `get_execution_result`
|
|
14
|
+
- **Tier 3 — Advanced Operations (5 tools):** `validated_create`, `validated_update`, `activate_workflow`, `get_session_metrics`, `server_health`
|
|
15
|
+
- **Tier 4 — Tool Parity (5 tools):** `test_workflow`, `server_autofix`, `partial_update_workflow`, `manage_executions`, `deploy_template`
|
|
16
|
+
- **Tier 5 — Workflow Generation (2 tools):** `plan_workflow`, `build_workflow`
|
|
17
|
+
- **Tier 6 — Credential Management (6 tools):** `list_credentials`, `get_credential_schema`, `create_credential`, `test_credential`, `update_credential`, `delete_credential`
|
|
18
|
+
- **Tier 7 — Config Intelligence (1 tool):** `collect_config`
|
|
19
|
+
- **Credential management** — full lifecycle: list, inspect schema (679 credential types), create, health-check, update, delete; 3-layer redaction prevents secrets in tool output
|
|
20
|
+
- **Workflow generation pipeline** — `plan_workflow` + `build_workflow` produce validated workflows from natural language; multi-pass auto-fix runs automatically
|
|
21
|
+
- **Config intelligence** — `collect_config` detects missing node configuration using 2,494 known gaps
|
|
22
|
+
- **validated_create / validated_update** — L1-L6 validation runs before any deploy; rejects invalid JSON before it reaches the n8n server
|
|
23
|
+
- **Diff-based updates** — `partial_update_workflow` applies node-level patches without replacing the entire workflow
|
|
24
|
+
- **K2 sandbox testing** — `test_workflow` validates workflow logic using mock credential responses (no real API calls required)
|
|
25
|
+
- **Session cost metrics** — every tool response includes token usage and cost; `get_session_metrics` returns cumulative session totals
|
|
26
|
+
- **Auto-fixer expanded** — 35 production fixers (up from 26), including 5 new L6 V2 fixers for flow integrity and an LLM fallback fixer
|
|
27
|
+
- **L5 safety guard** — prevents destructive auto-fixes on uncataloged community nodes
|
|
28
|
+
- **Catalog refreshed (2026-03-08)** — nodes: 1,545 (1,482 core + 63 community), credentials: 679, operations: 12,619, parameters: 38,005
|
|
29
|
+
- **Cursor and Windsurf support** — installation guide updated with setup instructions for both IDEs
|
|
30
|
+
- **Benchmark results (RAG-4.36)** — 287-observation study: p58 90% vs n8n-mcp 35% vs FlowEngine 35% vs no-MCP 25%; $0.25/prompt vs $0.76 (67% cheaper)
|
|
31
|
+
|
|
32
|
+
### Changed
|
|
33
|
+
|
|
34
|
+
- README completely rewritten to reflect v0.2.0 capabilities, fresh catalog counts, and competitive positioning
|
|
35
|
+
- TOOLS.md updated with all 34 tools, full input/output schemas, and configuration guide
|
|
36
|
+
- INSTALLATION.md updated with Cursor, Windsurf, and multi-environment setup
|
|
37
|
+
|
|
38
|
+
### Fixed
|
|
39
|
+
|
|
40
|
+
- `partial_update_workflow` connection format bug — n8n expects `{ main: [[targets]] }` wrapper (was crashing on all models)
|
|
41
|
+
- `server_autofix` JSON string input — now accepts workflow as JSON string or object
|
|
42
|
+
- n8n rejects empty `pinData: {}` — removed from workflow payload when empty
|
|
43
|
+
|
|
44
|
+
## [0.1.1] - 2026-02-06
|
|
45
|
+
|
|
46
|
+
### Fixed
|
|
47
|
+
|
|
48
|
+
- Bundle MCP server with esbuild to resolve private registry dependency (`@tsvika58/shared-utilities`) for `npx` installs
|
|
49
|
+
- Set npm package access to public (scoped packages default to private)
|
|
50
|
+
|
|
51
|
+
### Changed
|
|
52
|
+
|
|
53
|
+
- `bin` entry now points to bundled CJS file (`dist/mcp/server.bundle.cjs`) for universal compatibility
|
|
54
|
+
- Moved `@tsvika58/shared-utilities` from runtime to dev dependency (bundled into dist)
|
|
55
|
+
|
|
56
|
+
## [0.1.0] - 2026-02-06
|
|
57
|
+
|
|
58
|
+
### Added
|
|
59
|
+
|
|
60
|
+
- **MCP Server** with 8 tools for LLM integration via stdio transport
|
|
61
|
+
- `validate_workflow` — Full L1-L6 static validation pipeline
|
|
62
|
+
- `get_operation_schema` — Parameter requirements lookup for any node operation
|
|
63
|
+
- `check_parameter` — Real-time parameter validation against catalog
|
|
64
|
+
- `suggest_fix` — Concrete fix suggestions powered by 26 deterministic auto-fixers
|
|
65
|
+
- `list_operations` — Browse all operations for a given node type
|
|
66
|
+
- `list_nodes` — Search and browse 1,982 available n8n nodes
|
|
67
|
+
- `get_node_info` — Full node details, configuration, and metadata
|
|
68
|
+
- `find_similar_pattern` — Working examples from 9,178 real workflows
|
|
69
|
+
- **L1-L6 Static Validation Pipeline** (~1.4s total)
|
|
70
|
+
- L1: JSON structure, required fields, duplicate detection (~30ms)
|
|
71
|
+
- L2: Node type catalog validation (~60ms)
|
|
72
|
+
- L3: Credential type validation (~80ms)
|
|
73
|
+
- L4: Connection topology validation (~180ms)
|
|
74
|
+
- L5: Parameter configuration validation (~850ms)
|
|
75
|
+
- L6: Flow integrity, security, and expression patterns (~200ms)
|
|
76
|
+
- **AutoFix Infrastructure** with 26 deterministic fixers across L1-L6
|
|
77
|
+
- Zero LLM cost for automatic repairs
|
|
78
|
+
- 97%+ success rate on fixable errors
|
|
79
|
+
- **Node Catalog** with 1,982 nodes, 654 credential types, 40,179 operations
|
|
80
|
+
- **Pattern Library** with 9,178 real workflow patterns for discovery
|
|
81
|
+
- **Multi-client support** for Claude Desktop, Claude Code, and Gemini CLI
|
|
82
|
+
- **Drop-and-install** configuration via `AGENT_INSTALL.md`
|
|
83
|
+
- npm package published as `@path58/p58-n8n`
|
|
84
|
+
- ESM module support with shebang for direct `npx` execution
|
|
85
|
+
|
|
86
|
+
[0.2.0]: https://github.com/tsvika58/n8n-workflow-validator/releases/tag/v0.2.0
|
|
87
|
+
[0.1.1]: https://github.com/tsvika58/n8n-workflow-validator/releases/tag/v0.1.1
|
|
88
|
+
[0.1.0]: https://github.com/tsvika58/n8n-workflow-validator/releases/tag/v0.1.0
|
package/LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 Path58
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
package/README.md
ADDED
|
@@ -0,0 +1,281 @@
|
|
|
1
|
+
# p58-n8n — The n8n MCP Server That Gets Workflows Right
|
|
2
|
+
|
|
3
|
+
**Build n8n workflows from plain English. First try. Every time.**
|
|
4
|
+
|
|
5
|
+
[](https://www.npmjs.com/package/@path58/p58-n8n)
|
|
6
|
+
[](LICENSE)
|
|
7
|
+
[](https://github.com/tsvika58/n8n-workflow-validator/actions/workflows/ci.yml)
|
|
8
|
+
|
|
9
|
+
p58-n8n is an MCP server that gives your AI assistant deep knowledge of n8n — **1,545 nodes**, **12,619 operations**, **38,005 parameter rules** — so it can plan, build, validate, fix, and deploy workflows correctly. No manual iteration. No broken JSON.
|
|
10
|
+
|
|
11
|
+
---
|
|
12
|
+
|
|
13
|
+
## Quick Start (< 2 minutes)
|
|
14
|
+
|
|
15
|
+
### 1. Install
|
|
16
|
+
|
|
17
|
+
**Via npx (no install needed):**
|
|
18
|
+
```bash
|
|
19
|
+
npx -y @path58/p58-n8n
|
|
20
|
+
```
|
|
21
|
+
|
|
22
|
+
**Or install globally:**
|
|
23
|
+
```bash
|
|
24
|
+
npm install -g @path58/p58-n8n
|
|
25
|
+
```
|
|
26
|
+
|
|
27
|
+
### 2. Configure your AI client
|
|
28
|
+
|
|
29
|
+
**Claude Code** (one command):
|
|
30
|
+
```bash
|
|
31
|
+
claude mcp add p58-n8n -- npx -y @path58/p58-n8n
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
**Claude Desktop** — add to `claude_desktop_config.json`:
|
|
35
|
+
```json
|
|
36
|
+
{
|
|
37
|
+
"mcpServers": {
|
|
38
|
+
"p58-n8n": {
|
|
39
|
+
"command": "npx",
|
|
40
|
+
"args": ["-y", "@path58/p58-n8n"]
|
|
41
|
+
}
|
|
42
|
+
}
|
|
43
|
+
}
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
Config file location: `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) or `%APPDATA%\Claude\claude_desktop_config.json` (Windows). Restart Claude Desktop after saving.
|
|
47
|
+
|
|
48
|
+
**Cursor / Windsurf** — see [docs/INSTALLATION.md](docs/INSTALLATION.md) for full multi-client setup.
|
|
49
|
+
|
|
50
|
+
### 3. Try it
|
|
51
|
+
|
|
52
|
+
```
|
|
53
|
+
Build me a workflow that monitors a Google Sheet and sends a Slack message
|
|
54
|
+
when a new row is added. Use the Google Sheets and Slack credentials I already
|
|
55
|
+
have configured in n8n.
|
|
56
|
+
```
|
|
57
|
+
|
|
58
|
+
p58-n8n will plan the workflow, look up the exact node parameters, validate the JSON against L1-L6 rules, wire your existing credentials, and deploy it to your n8n instance — in one conversation.
|
|
59
|
+
|
|
60
|
+
---
|
|
61
|
+
|
|
62
|
+
## What Makes p58 Different
|
|
63
|
+
|
|
64
|
+
### Benchmark: 90% success rate on real-world workflows
|
|
65
|
+
|
|
66
|
+
| | **p58-n8n** | **n8n-mcp** | **FlowEngine** | **Claude alone** |
|
|
67
|
+
|---|---|---|---|---|
|
|
68
|
+
| **Workflow success rate** | **90%** | 35% | 35% | 25% |
|
|
69
|
+
| **Cost per prompt** | **$0.25** | $0.76 | — | ~$0.05 |
|
|
70
|
+
| **Credential management** | ✅ Full CRUD | ❌ | ❌ | ❌ |
|
|
71
|
+
| **Config intelligence** | ✅ 2,494 known gaps | ❌ | ❌ | ❌ |
|
|
72
|
+
|
|
73
|
+
*Benchmarked on 287 real-world workflow generation tasks across Sonnet, Opus, and Haiku models.*
|
|
74
|
+
|
|
75
|
+
### The full pipeline — not just JSON generation
|
|
76
|
+
|
|
77
|
+
Most tools generate workflow JSON and stop. p58-n8n runs the complete cycle:
|
|
78
|
+
|
|
79
|
+
```
|
|
80
|
+
plan_workflow → build_workflow → validate (L1-L6) → server_autofix →
|
|
81
|
+
wire credentials → collect_config → validated_create → execute_workflow
|
|
82
|
+
```
|
|
83
|
+
|
|
84
|
+
Every step is backed by structured catalog data. No guessing.
|
|
85
|
+
|
|
86
|
+
### Deep validation — 6 levels, not vibes
|
|
87
|
+
|
|
88
|
+
```
|
|
89
|
+
L1 Structure → L2 Nodes → L3 Credentials → L4 Connections → L5 Parameters → L6 Patterns
|
|
90
|
+
~30ms ~60ms ~80ms ~180ms ~850ms ~200ms
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
L5 alone checks 38,005 parameter rules. L6 catches dead ends, retry loops without error handlers, and hardcoded secrets. Combined: ~1.4 seconds for a full audit.
|
|
94
|
+
|
|
95
|
+
### 35 auto-fixers — not a prompt
|
|
96
|
+
|
|
97
|
+
When validation fails, p58-n8n applies deterministic fixes: reconnect orphaned nodes, add missing credentials, patch wrong parameter types, rebuild broken flow paths. 97%+ success rate. Zero LLM cost.
|
|
98
|
+
|
|
99
|
+
### Credential management — a category no competitor touches
|
|
100
|
+
|
|
101
|
+
```
|
|
102
|
+
list_credentials → get_credential_schema → create_credential →
|
|
103
|
+
test_credential → update_credential → delete_credential
|
|
104
|
+
```
|
|
105
|
+
|
|
106
|
+
The AI can discover what credentials your n8n instance has, look up what fields any of 679 credential types need, create or update credentials, health-check them, and wire them into new workflows — all without leaving the conversation.
|
|
107
|
+
|
|
108
|
+
### 67% cheaper than the alternative
|
|
109
|
+
|
|
110
|
+
p58-n8n uses 80% less token context than n8n-mcp by serving structured catalog data instead of documentation blobs. At scale: $0.25/prompt vs $0.76 for czlonkowski.
|
|
111
|
+
|
|
112
|
+
---
|
|
113
|
+
|
|
114
|
+
## 34 Tools in 7 Tiers
|
|
115
|
+
|
|
116
|
+
### Tier 1 — Validation & Analysis
|
|
117
|
+
|
|
118
|
+
| Tool | What it does |
|
|
119
|
+
|------|-------------|
|
|
120
|
+
| `validate_workflow` | Full L1-L6 validation — structure, nodes, credentials, connections, parameters, flow patterns |
|
|
121
|
+
| `get_operation_schema` | Exact parameter requirements for any node operation (ground truth, not LLM guesses) |
|
|
122
|
+
| `check_parameter` | Real-time parameter validation — type checking, enum validation, typo detection |
|
|
123
|
+
| `suggest_fix` | Concrete fix suggestions for validation issues — rename, add, remove, or replace |
|
|
124
|
+
| `list_nodes` | Browse and search 1,545 n8n nodes by category or keyword |
|
|
125
|
+
| `list_operations` | All operations for a node type, grouped by resource |
|
|
126
|
+
| `get_node_info` | Full node details — operations, credentials, properties, version |
|
|
127
|
+
| `find_similar_pattern` | Search 10,003 real workflows for working examples |
|
|
128
|
+
|
|
129
|
+
### Tier 2 — Server CRUD
|
|
130
|
+
|
|
131
|
+
| Tool | What it does |
|
|
132
|
+
|------|-------------|
|
|
133
|
+
| `get_workflow` | Fetch a workflow from your n8n instance by ID |
|
|
134
|
+
| `list_workflows` | List all workflows with filtering and pagination |
|
|
135
|
+
| `create_workflow` | Create a workflow on your n8n server |
|
|
136
|
+
| `update_workflow` | Replace a workflow entirely |
|
|
137
|
+
| `delete_workflow` | Delete a workflow (safe — checks for dependents) |
|
|
138
|
+
| `execute_workflow` | Trigger workflow execution |
|
|
139
|
+
| `get_execution_result` | Fetch the result of a completed execution |
|
|
140
|
+
|
|
141
|
+
### Tier 3 — Advanced Operations
|
|
142
|
+
|
|
143
|
+
| Tool | What it does |
|
|
144
|
+
|------|-------------|
|
|
145
|
+
| `validated_create` | Validate + deploy in one step — rejects invalid JSON before it hits your server |
|
|
146
|
+
| `validated_update` | Validate + update in one step |
|
|
147
|
+
| `activate_workflow` | Toggle workflow active/inactive state |
|
|
148
|
+
| `get_session_metrics` | Token usage and cost tracking for the current session |
|
|
149
|
+
| `server_health` | Check your n8n instance connectivity and status |
|
|
150
|
+
|
|
151
|
+
### Tier 4 — Tool Parity
|
|
152
|
+
|
|
153
|
+
| Tool | What it does |
|
|
154
|
+
|------|-------------|
|
|
155
|
+
| `test_workflow` | Sandbox-test a workflow using K2 mock responses (no real credentials needed) |
|
|
156
|
+
| `server_autofix` | Run all 35 auto-fixers on a workflow and return the repaired JSON |
|
|
157
|
+
| `partial_update_workflow` | Diff-based partial updates — change one node without replacing the whole workflow |
|
|
158
|
+
| `manage_executions` | List, filter, and delete workflow execution history |
|
|
159
|
+
| `deploy_template` | Deploy a pre-built template from the workflow library |
|
|
160
|
+
|
|
161
|
+
### Tier 5 — Workflow Generation
|
|
162
|
+
|
|
163
|
+
| Tool | What it does |
|
|
164
|
+
|------|-------------|
|
|
165
|
+
| `plan_workflow` | Generate a structured workflow plan from a natural language description |
|
|
166
|
+
| `build_workflow` | Build a complete workflow JSON from a plan — uses catalog data for correct parameters |
|
|
167
|
+
|
|
168
|
+
### Tier 6 — Credential Management
|
|
169
|
+
|
|
170
|
+
| Tool | What it does |
|
|
171
|
+
|------|-------------|
|
|
172
|
+
| `list_credentials` | List n8n server credentials with optional schema enrichment |
|
|
173
|
+
| `get_credential_schema` | Required and optional fields for any of 679 credential types (fuzzy match) |
|
|
174
|
+
| `create_credential` | Create credentials with 5-step validation |
|
|
175
|
+
| `test_credential` | Health-check probe — verifies connectivity for top credential types |
|
|
176
|
+
| `update_credential` | Partial update — GET → merge → PATCH (preserves unmodified fields) |
|
|
177
|
+
| `delete_credential` | Safe deletion with dependent workflow scan |
|
|
178
|
+
|
|
179
|
+
### Tier 7 — Config Intelligence
|
|
180
|
+
|
|
181
|
+
| Tool | What it does |
|
|
182
|
+
|------|-------------|
|
|
183
|
+
| `collect_config` | Detect missing node configuration from your live n8n instance (covers 2,494 known gaps) |
|
|
184
|
+
|
|
185
|
+
Full reference with input/output schemas: [docs/TOOLS.md](docs/TOOLS.md)
|
|
186
|
+
|
|
187
|
+
---
|
|
188
|
+
|
|
189
|
+
## Examples
|
|
190
|
+
|
|
191
|
+
### Build a Slack notification workflow
|
|
192
|
+
|
|
193
|
+
```
|
|
194
|
+
Plan and build me a workflow that triggers every day at 9am, queries a PostgreSQL
|
|
195
|
+
database for orders placed in the last 24 hours, and posts a summary to the
|
|
196
|
+
#daily-reports Slack channel.
|
|
197
|
+
```
|
|
198
|
+
|
|
199
|
+
p58-n8n will: call `plan_workflow` → `build_workflow` → `validate_workflow` (L1-L6) → `list_credentials` to find your Slack and PostgreSQL credentials → `validated_create` to deploy to n8n.
|
|
200
|
+
|
|
201
|
+
### Fix a broken workflow
|
|
202
|
+
|
|
203
|
+
```
|
|
204
|
+
I imported this workflow from n8n.io but it has errors. Can you validate and fix it?
|
|
205
|
+
[paste workflow JSON]
|
|
206
|
+
```
|
|
207
|
+
|
|
208
|
+
p58-n8n will: call `validate_workflow` → identify L1-L6 issues → call `server_autofix` to apply all 35 fixers → return the repaired workflow with a diff of what changed.
|
|
209
|
+
|
|
210
|
+
### Set up credentials and deploy
|
|
211
|
+
|
|
212
|
+
```
|
|
213
|
+
I need to create a Slack credential and then use it in a new workflow.
|
|
214
|
+
My Slack bot token is in my clipboard.
|
|
215
|
+
```
|
|
216
|
+
|
|
217
|
+
p58-n8n will: call `get_credential_schema` for Slack → `create_credential` with validation → `test_credential` to verify → wire it into the workflow → deploy.
|
|
218
|
+
|
|
219
|
+
---
|
|
220
|
+
|
|
221
|
+
## Installation
|
|
222
|
+
|
|
223
|
+
See [docs/INSTALLATION.md](docs/INSTALLATION.md) for full setup instructions:
|
|
224
|
+
|
|
225
|
+
- Claude Desktop (macOS, Windows, Linux)
|
|
226
|
+
- Claude Code (one-command setup)
|
|
227
|
+
- Cursor and Windsurf
|
|
228
|
+
- Global install vs npx
|
|
229
|
+
- Troubleshooting
|
|
230
|
+
|
|
231
|
+
---
|
|
232
|
+
|
|
233
|
+
## Benchmarks
|
|
234
|
+
|
|
235
|
+
Benchmarked on 287 real-world workflow generation tasks using prompts from the n8n community. Tests ran across three models (Haiku 4.5, Sonnet 4.6, Opus 4.6) with four configurations (p58-n8n, n8n-mcp, FlowEngine, and no MCP server).
|
|
236
|
+
|
|
237
|
+
| Metric | p58-n8n | n8n-mcp | FlowEngine | No MCP |
|
|
238
|
+
|--------|---------|---------|------------|--------|
|
|
239
|
+
| **Overall success rate** | **90%** | 35% | 35% | 25% |
|
|
240
|
+
| **Sonnet 4.6** | 80% | — | — | — |
|
|
241
|
+
| **Opus 4.6** | 80% | — | — | — |
|
|
242
|
+
| **Haiku 4.5** | 60% | — | — | — |
|
|
243
|
+
| **Cost per prompt** | $0.25 | $0.76 | — | ~$0.05 |
|
|
244
|
+
| **Credential prompts** | **78%** | 0% | 0% | 0% |
|
|
245
|
+
|
|
246
|
+
*"Credential prompts" = workflows requiring credential setup that competitors cannot attempt at all.*
|
|
247
|
+
|
|
248
|
+
---
|
|
249
|
+
|
|
250
|
+
## License
|
|
251
|
+
|
|
252
|
+
MIT — see [LICENSE](LICENSE) for details.
|
|
253
|
+
|
|
254
|
+
**Built by [Path58](https://path58.com)**
|
|
255
|
+
|
|
256
|
+
---
|
|
257
|
+
|
|
258
|
+
## Contributing
|
|
259
|
+
|
|
260
|
+
p58-n8n is in **soft launch** (friends & family). Issues and feedback welcome:
|
|
261
|
+
|
|
262
|
+
- **Bugs:** [GitHub Issues](https://github.com/tsvika58/n8n-workflow-validator/issues)
|
|
263
|
+
- **Email:** tvagman@gmail.com
|
|
264
|
+
|
|
265
|
+
---
|
|
266
|
+
|
|
267
|
+
## Architecture
|
|
268
|
+
|
|
269
|
+
p58-n8n runs as a local stdio MCP server. No cloud services required for basic use.
|
|
270
|
+
|
|
271
|
+
```
|
|
272
|
+
AI Client (Claude / Cursor / Windsurf)
|
|
273
|
+
↕ MCP Protocol (stdio)
|
|
274
|
+
p58-n8n Server
|
|
275
|
+
├── Validation Engine (L1-L6, ~1.4s)
|
|
276
|
+
├── AutoFix Engine (35 fixers, 97%+ success)
|
|
277
|
+
├── Node Catalog (1,545 nodes, 12,619 ops, 38,005 params)
|
|
278
|
+
└── n8n API Client (deploy, execute, credential management)
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
**Prerequisites:** Node.js 18+ and a running n8n instance (for server CRUD and credential tools; validation and catalog tools work offline).
|