Index of every page on this site: llms.txt
All skills
Run it yourself

Build repeatable AI agents and no-code automations with n8n and Zapier: pick the right tool, write the full agent prompt, set up self-hosted n8n, and ship a tested workflow with error handling, cost guards, and a security checklist. Use when automating a repetitive task, building a research or inbox-triage agent, or choosing between n8n, Zapier, and other automation tools.

  • 4 files
  • 25.4 KB
  • Updated 5 days ago
  • GitHub
Use this Skill: https://skilld.dev/gh/jokerhtml/needcraft-skills/ai-agents-automation

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

examplesresearch-agent-build.md

≈863 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Worked build: daily research brief agent (n8n)

Status: build guide (example). The steps below are a complete build you can follow. The sample output is illustrative — it was written as an example, not produced by a run. Run the workflow yourself, then replace this note with your run date and result.

The job

Every morning, produce a short brief on one topic: what changed, what the sources say, what to do next.

Build (n8n)

  1. Schedule Trigger — daily, 07:00.
  2. HTTP Request — GET the source page or RSS feed for your topic. Example source: the n8n changelog, https://docs.n8n.io/release-notes/ (check the URL is current before use).
  3. AI Agent — model of your choice; prompt below; input = text from step 2.
  4. IF node — continue only when the output is non-empty and contains at least one source link.
  5. Gmail / Slack / Google Docs — deliver the brief to one destination you actually read.

The agent prompt (filled 8-part block)

  1. Title: Daily research brief — n8n changelog.

  2. Copy-ready prompt:

    You are a research assistant. Context: I follow n8n product changes.
    Task: read the sources below and write a brief under 200 words with
    three sections: What changed, What the sources say, What to do next.
    Inputs: [paste the fetched changelog text].
    Constraints: use only claims present in the sources; name the source
    after each claim; if a claim has no source, write "unverified".
    Output structure: three headings, bullets under each, one source link
    per bullet. Quality checks: flag any claim you cannot support from the
    sources and remove it.
  3. Placeholders: [paste the fetched changelog text] = the raw text from the HTTP Request node. No other placeholders.

  4. Recommended tool/model: any chat model on your plan; verify the model name is current before publishing.

  5. How to use: open the workflow, run it once manually, read the brief in your chosen destination.

  6. Expected output (example, not a real run):

    ## What changed
    - n8n released a new AI Agent node update. Source: n8n changelog.
    
    ## What the sources say
    - The update adds error handling for tool calls. Source: n8n changelog.
    
    ## What to do next
    - Read the changelog entry and test the new node in a copy of your workflow.

    The lines above are illustrative wording, not claims about any real release. Never present them as release notes.

  7. Iteration prompt: "Check every factual claim in your answer against the sources I provided. List any claim you cannot support, and remove it."

  8. Limitations: the model can hallucinate changelog entries; open every link before sharing the brief.

Cost check (before you switch it on)

  • n8n Cloud Starter allows 2,500 executions/month (last checked 16 Nov 2026). One run a day = ~30 executions/month. The AI step's cost depends on your model provider — set a budget cap there first.
  • On Zapier Free (100 tasks/month), this build as a two-step Zap is not possible as written; the multi-step version needs a paid plan.

Test run record

Date Input Output Result
(fill in after your first run)

No-code alternative (Zapier)

Schedule by Zapier → RSS by Zapier → AI by Zapier → Gmail. Free-plan note: two-step Zaps only, so split it into two Zaps with a storage step between them, or use a paid plan.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 90f40b1. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 4 days ago.

Activeupdated 5 days ago
Other metadata
metadata
{
  "version": "1.0.0",
  "date": "2026-11-16",
  "author": "Muhammad Saad Ranjha / NeedCraft"
}

README badge

README badge for jokerhtml/needcraft-skills/ai-agents-automation