From atum-nocode
Build, debug, and optimize Make.com scenarios. Generates blueprints from natural language, diagnoses execution failures, optimizes operation consumption, and manages error handlers. Use when asked to create a Make scenario, fix a Make error, optimize operations, generate a blueprint, or automate a workflow with Make.com.
npx claudepluginhub arnwaldn/atum-plugins-collection --plugin atum-nocodeThis skill uses the workspace's default tool permissions.
Expert en construction, debug et optimisation de scénarios Make.com.
Executes pre-written implementation plans: critically reviews, follows bite-sized steps exactly, runs verifications, tracks progress with checkpoints, uses git worktrees, stops on blockers.
Guides idea refinement into designs: explores context, asks questions one-by-one, proposes approaches, presents sections for approval, writes/review specs before coding.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Expert en construction, debug et optimisation de scénarios Make.com.
Convert a plain-language description into a Make.com blueprint JSON structure.
Process:
Blueprint Structure Reference:
{
"name": "Scenario Name",
"flow": [
{
"id": 1,
"module": "gateway:CustomWebHook",
"mapper": {},
"metadata": { "designer": { "x": 0, "y": 0 } }
}
],
"metadata": { "version": 1 }
}
When a scenario fails:
mcp__claude_ai_Make__executions_list, mcp__claude_ai_Make__executions_get-detail)Reduce operation consumption:
map(), get(), slice() in mapping instead of Iterator + separate module{{ifempty(1.data; "stop")}} to skip empty runs| Pattern | Use Case |
|---|---|
| Resume | Ignore error, continue flow (non-critical modules) |
| Break | Stop execution, retry later (rate limits, temp failures) |
| Rollback | Undo all operations in transaction (payment flows) |
| Commit | Save partial results, stop (batch processing) |
Best practice: Every scenario should have at least one error handler on the most critical module (usually the first API call or database write).
Webhook → Search Airtable (deduplicate) → Router
├─ [exists] → Update Record
└─ [new] → Create Record
HTTP Request → Error Handler (Break)
└─ Retry: 3 times, 60s intervals
Trigger → Iterator (split array)
→ Parallel module calls
→ Array Aggregator (recombine)
→ Final action
Webhook → Data Store (queue) → Respond immediately
Scheduler (every 5min) → Search Data Store → Iterator → Process → Delete from store
Use these Make.com MCP tools for direct scenario management:
mcp__claude_ai_Make__scenarios_list — List all scenariosmcp__claude_ai_Make__scenarios_get — Get scenario detailsmcp__claude_ai_Make__scenarios_create — Create new scenariomcp__claude_ai_Make__scenarios_update — Update scenariomcp__claude_ai_Make__scenarios_activate / deactivatemcp__claude_ai_Make__scenarios_run — Execute manuallymcp__claude_ai_Make__executions_list / get / get-detail — Debugmcp__claude_ai_Make__connections_list / get — Check connectionsmcp__claude_ai_Make__hooks_* — Manage webhooksmcp__claude_ai_Make__data-stores_* — Manage data stores