chore: sync skills pack v1.22.0 (validator guidance aligned with 2.63.0)

[czlonkowski]

Summary

Syncs data/skills/ from n8n-skills v1.22.0 (czlonkowski/n8n-skills#36), which aligns all validator guidance with the 2.63.0 validator overhaul (#911):

• No more fabricated "known issues Validation False Positives: Google Sheets node reports errors for working configurations #304/validator cannot properly validate setNode with broken JSON definition inside it #306/False positive: Template literal validation incorrectly flags working expressions #338" or stale "ai-friendly reduces false positives by 60%" claims — the pack now documents the real profile gating (minimal/runtime = errors + security/deprecation; ai-friendly/strict add best-practice advisories).
• Precise template-literal guidance (backticks with ${} interpolation are valid inside {{ }}), optional chaining and bracket access documented as valid.
• Code-node guidance: bare-object returns auto-wrap in All Items mode; obsolete validator-appeasement advice removed.
• Every rewritten behavioral claim in the source pack was verified empirically against the 2.63.0 validator build under all four profiles.

Also brings in every skill added to the pack since the last sync (May): agents, error-handling, binary-and-data, code-tool, subworkflows, multi-instance, self-hosting, and the using-n8n-mcp-skills router — these ship in the npm/Docker artifacts and are served via MCP Resources.

sync-skills.ts now skips skill-creator eval workspace directories (skills/*-workspace/) so local eval debris can never ship in artifacts.

Tests

• tests/unit/mcp/skills/registry.test.ts: 17/17 pass against the synced tree.
• Content gate: zero client/internal references in the synced markdown.

Concieved by Romuald Członkowski - www.aiadvisors.pl/en

🤖 Generated with Claude Code

[Copilot]

Pull request overview

This PR syncs the repository’s bundled data/skills/ content to n8n-skills v1.22.0 (including multiple newly added skills) and updates guidance to reflect the validator profile gating behavior introduced in the n8n-mcp 2.63.0 validator overhaul. It also hardens the sync script to avoid shipping local “eval workspace” debris.

Changes:

• Sync data/skills/ to the latest upstream pack (new skills + widespread guidance refresh, especially validator/profile behavior and expression validity).
• Update workflow/tooling guidance in several skills (validation profiles, workflow retrieval modes, expression/template-literal notes, performance notes, etc.).
• Update scripts/sync-skills.ts to skip copying *-workspace directories.

Reviewed changes

Copilot reviewed 76 out of 76 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
scripts/sync-skills.ts	Skips copying *-workspace directories during skills sync to prevent local eval debris from shipping.
data/skills/using-n8n-mcp-skills/README.md	Adds router skill documentation describing pack routing/activation and tool surface overview.
data/skills/n8n-workflow-patterns/SKILL.md	Expands description and adds lifecycle/performance guidance; updates AI agent guide reference wording.
data/skills/n8n-workflow-patterns/scheduled_tasks.md	Adds timezone clarification for schedule trigger hour-based fields.
data/skills/n8n-workflow-patterns/README.md	Updates skill metadata listing (file counts/lines).
data/skills/n8n-workflow-patterns/database_operations.md	Adds guidance about alwaysOutputData for DB write operations to prevent downstream “0 items” breaks.
data/skills/n8n-validation-expert/README.md	Updates validation profile definitions and removes stale/incorrect false-positive claims.
data/skills/n8n-subworkflows/SUBWORKFLOW_PATTERNS.md	Adds detailed sub-workflow patterns (mode all vs each, N+1 split, fire-and-forget parallelization).
data/skills/n8n-subworkflows/README.md	Introduces the sub-workflows skill and its key concepts/structure.
data/skills/n8n-subworkflows/NAMING_AND_DISCOVERY.md	Adds naming/discovery guidance for sub-workflows under MCP constraints.
data/skills/n8n-self-hosting/SKILL.md	Introduces end-to-end self-hosting guidance (single vs queue mode, secrets, verification, handoff).
data/skills/n8n-self-hosting/SINGLE_MODE.md	Adds single-mode deployment notes, DB options, migration, verification checklist.
data/skills/n8n-self-hosting/SECURITY.md	Adds secret generation + hardening checklist guidance for self-hosted deployments.
data/skills/n8n-self-hosting/README.md	Adds overview/metadata for the self-hosting skill.
data/skills/n8n-self-hosting/QUEUE_MODE.md	Adds queue-mode architecture and scaling guidance.
data/skills/n8n-self-hosting/DAY2.md	Adds day-2 ops guidance (update/backup/restore) for self-hosted deployments.
data/skills/n8n-node-configuration/README.md	Updates metadata listing (file counts/lines).
data/skills/n8n-node-configuration/DEPENDENCIES.md	Adds a “quick reference” section explaining displayOptions dependency patterns and discovery workflow.
data/skills/n8n-multi-instance/README.md	Introduces multi-instance targeting/switching/verification guidance for n8n_instances.
data/skills/n8n-mcp-tools-expert/WORKFLOW_GUIDE.md	Updates workflow retrieval mode guidance (draft vs active, adds filtered guidance) and removes obsolete generation-tool section.
data/skills/n8n-mcp-tools-expert/README.md	Updates metadata listing (file counts/lines).
data/skills/n8n-mcp-tools-expert/OPERATIONS_GUIDE.md	Adds reference guide for templates, data tables, and “self-help” tooling documentation.
data/skills/n8n-expression-syntax/SKILL.md	Adds transform gatekeeper guidance and explicit performance note on expression complexity.
data/skills/n8n-expression-syntax/README.md	Updates metadata listing (file counts/lines).
data/skills/n8n-expression-syntax/COMMON_MISTAKES.md	Updates template-literal guidance to reflect valid usage inside {{ }} and clarifies failure mode outside expressions.
data/skills/n8n-error-handling/README.md	Introduces error-handling skill and structured guidance for routed failures/retries.
data/skills/n8n-error-handling/NODE_ERROR_OUTPUTS.md	Adds detailed per-node error output wiring and verification guidance.
data/skills/n8n-error-handling/ERROR_WORKFLOWS.md	Adds workflow-level error workflow guidance and caveats.
data/skills/n8n-code-tool/README.md	Introduces Code Tool (LangChain tool) guidance and contrasts with Code node.
data/skills/n8n-code-tool/INPUT_SCHEMA.md	Adds structured-input schema guidance for Code Tool.
data/skills/n8n-code-tool/ERROR_PATTERNS.md	Adds common Code Tool error patterns with fixes.
data/skills/n8n-code-python/STANDARD_LIBRARY.md	Adds a concise quick-reference snippet section for common stdlib usage.
data/skills/n8n-code-python/README.md	Updates guidance for JS HTTP helpers (this.helpers.httpRequest) and metadata listing.
data/skills/n8n-code-python/ERROR_PATTERNS.md	Updates JS HTTP helper example to this.helpers.httpRequest.
data/skills/n8n-code-python/COMMON_PATTERNS.md	Adds “Quick Pattern Snippets” section with condensed copy-ready patterns.
data/skills/n8n-code-javascript/README.md	Updates helper guidance to prefer this.helpers.httpRequest and clarifies return-shape behavior.
data/skills/n8n-code-javascript/DATA_ACCESS.md	Updates helper usage to this.helpers.httpRequest and adds performance guidance.
data/skills/n8n-code-javascript/COMMON_PATTERNS.md	Adds best-practices appendix section.
data/skills/n8n-code-javascript/BUILTIN_FUNCTIONS.md	Updates HTTP helper docs to this.helpers.httpRequest and clarifies task-runner sandbox constraints.
data/skills/n8n-binary-and-data/README.md	Introduces binary/data handling guidance and the agent-tool JSON-only boundary.
data/skills/n8n-binary-and-data/MERGE_FOR_CONTEXT.md	Adds merge-by-position pattern to reattach binary after JSON-only transforms.
data/skills/n8n-binary-and-data/CDN_REQUIREMENT.md	Adds guidance on URL/CDN requirements for rendering binary in chat surfaces.
data/skills/n8n-binary-and-data/BINARY_BASICS.md	Adds in-depth binary slot anatomy and read/write/verification guidance.
data/skills/n8n-agents/SYSTEM_PROMPT.md	Adds system-prompt design guidance for agent behavior/tool modularization.
data/skills/n8n-agents/SUBWORKFLOW_AS_TOOL.md	Adds guidance on exposing sub-workflows as agent tools via Tool Workflow nodes.
data/skills/n8n-agents/STRUCTURED_OUTPUT.md	Adds structured output parser guidance with retry/autoFix and fixer model wiring.
data/skills/n8n-agents/README.md	Introduces the agents skill and its scope/structure.
data/skills/n8n-agents/RAG.md	Adds a thin RAG overview and “rule out cheaper lookups” guidance.
data/skills/n8n-agents/MEMORY.md	Adds memory type selection and session key handling guidance.
data/skills/n8n-agents/HUMAN_REVIEW.md	Adds HITL (human review) gating patterns for tool safety and approval UX.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[czlonkowski]

This was checked empirically just now: a minimal compose file with deploy.replicas: 2 and plain docker compose up -d (no Swarm, Docker Compose v5.3.0) starts two containers — probe-1 and probe-2 both running. Ignoring the deploy section outside Swarm was legacy Compose v1 behavior; Compose v2+ implements the Compose Spec, where deploy.replicas is honored by compose up (see https://docs.docker.com/reference/compose-file/deploy/#replicas). The doc anticipates this exact confusion — it explicitly says v2 honors it '(here it is not Swarm-only)' and documents --scale including its precedence over replicas. No change needed.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[github-actions]

Test Results Summary

📊 Artifacts

• 📄 Test Results
• 📊 Coverage Report
• ⚡ Benchmark Results

---

Generated at Fri, 03 Jul 2026 17:32:27 GMT
Commit: 9a2f57b
Run: #1281