n8nauto
Get Started

Use skills In n8n agent node

This template gives you a framework to use skills in any n8n agent. You can use this as a starting point and add any other tools or patterns needed for your use case. What are “skills”? Skills are a context management standard created by Anthropic for use in Claude code. Basically, instead of having a HUGE system prompt, skills split that into lots of small, structured instruction files that tell an agent how to do a specific kind of task. Instead of stuffing a massive prompt full of rules, the agent: finds the relevant skill file reads it and follows the steps inside It’s a simple pattern that makes managing system prompts for general purpose agents much more straightforward. See an example of a skills repo here. What this workflow does Responds to messages in n8n Chat (or Chat Hub) Builds an “available skills” index from one or more GitHub repos Lets the agent browse folders + fetch skill files (Markdown) as needed Uses the skill content to guide how it completes tasks How it (roughly) works A chat message comes in. The workflow lists directories in the configured skills repos (root + if present), filters out noise, and merges everything into one directory map. That directory map gets injected into the agent’s system prompt so it knows what skill files exist. When it needs instructions, it uses tools: List Files by Path Name to explore folders * Get a File From GitHub to pull the skill file as raw text (no base64) Same “skills” pattern as the CLI tools The flow is: find a skill → open it → follow the steps, the same way it works in the CLI tools, but running inside n8n, so you don’t need to download or install anything locally. How to set it up (Required) Add your GitHub credentials to each node that needs it (Required) Add your OpenRouter credentials to the chat node or replace with the provider of your choosing (Optional) Add more repos to (any skills GitHub repo works as long as your credentials have access to it, such as any public repo) (Optional) Add more tools and turn it into whatever agent you actually need
21 nodesmanual trigger54 views0 copiesProductivity
ChatTriggerAgentGithubSplitOutMemoryBufferWindowLmChatOpenRouterFilterGithubTool

Workflow JSON

{"meta":{"instanceId":"8b2425319537dba5a3f3f6bd0c9a64e19c8e987187f834687b2771d8254cc309","templateCredsSetupCompleted":true},"nodes":[{"id":"2f22da99-20e2-413d-aff2-22342ab555f3","name":"When chat message received","type":"@n8n/n8n-nodes-langchain.chatTrigger","position":[-960,0],"webhookId":"3611645b-0df7-4c25-8d3b-f4abb9b01fb7","parameters":{"options":{}},"typeVersion":1.4},{"id":"f48e987d-c929-4cb6-926a-898864202399","name":"AI Agent","type":"@n8n/n8n-nodes-langchain.agent","position":[544,0],"parameters":{"text":"={{ $('When chat message received').item.json.chatInput }}","options":{"maxIterations":150,"systemMessage":"=You are a helpful assistant. Your capabilities and instructions for completing tasks are stored in skill files in skill directories, whic his listed below in the Available Skills section\n\n## How You Work\nYou do not have built-in knowledge of how to complete tasks. Your instructions live in the skill files. When a user asks you to do something:\n\n1. Identify which skill(s) are relevant to the request, traversing directories using the `List Files by Path Name` tool as needed (ideally concurrently)\n2. Fetch those skill files (use concurrent tool calls when fetching multiple)\n3. Follow the instructions in those files to complete the task\n\n## Available Skills Files and Directories\n{{ \n// 🚨 ONLY UPDATE THIS IF YOU KNOW WHAT YOU'RE DOING 🚨\nJSON.stringify(\n  $('Merge Directory Structures')\n    .all()\n    .map(item => ({\n      \"type\": item.json.type,\n      \"path\": item.json.path,\n      \"orgName\": item.json.url.split('/')[4],\n      \"repoName\": item.json.url.split('/')[5],\n    }))\n    // this removes any files that we don't want to\n    // use up context displaying\n    .filter(i => \n      i.path.toLowerCase().split('/').last() !== \"readme.md\" && \n      i.path.toLowerCase().split('/').last() !== \"template\" &&\n      i.path.toLowerCase().split('/').last() !== \"license\"\n    ),\n  null,\n  2\n) \n}}\n\n## Important\n- Do NOT answer based on general knowledge — your instructions are in the skill files"},"promptType":"define"},"executeOnce":true,"typeVersion":3.1},{"id":"3813ec64-e25d-4f3a-9f63-667f28d2d75b","name":"List Root Dirs","type":"n8n-nodes-base.github","position":[-128,-80],"webhookId":"4ec37c56-9985-42e6-baf5-fe4ac627182b","parameters":{"owner":{"__rl":true,"mode":"url","value":"={{ $json.SKILLS_REPOS }}"},"filePath":"/","resource":"file","operation":"list","repository":{"__rl":true,"mode":"url","value":"={{ $json.SKILLS_REPOS }}"}},"credentials":{"githubApi":{"id":"aChEhZC5dvKq49tr","name":"liamdmcgarrigle read only"}},"typeVersion":1.1},{"id":"2bf04f87-8672-426a-af41-89de278c4f35","name":"Set GitHub Repo URLs","type":"n8n-nodes-base.set","position":[-608,0],"parameters":{"options":{},"assignments":{"assignments":[{"id":"8955dd58-72aa-49a9-a6aa-7c1679ee5b51","name":"SKILLS_REPOS","type":"array","value":"=[\n  \"https://github.com/anthropics/skills\",\n  \"https://github.com/anthropics/knowledge-work-plugins\"\n]"}]}},"typeVersion":3.4},{"id":"8533f02e-15b2-46e4-8324-8f0b97c438be","name":"Split Out","type":"n8n-nodes-base.splitOut","position":[-320,0],"parameters":{"options":{},"fieldToSplitOut":"SKILLS_REPOS"},"typeVersion":1},{"id":"36b931b8-6bde-4c3c-af00-7d0f41338e3a","name":"Sticky Note1","type":"n8n-nodes-base.stickyNote","position":[1088,384],"parameters":{"color":7,"height":288,"content":"HTTP request is used here instead of the built in GitHub node because the node returns base 64, but in the HTTP node we can request it as a text file through headers"},"typeVersion":1},{"id":"3fc14e56-09c1-4450-98d3-aadcc00d20af","name":"Sticky Note2","type":"n8n-nodes-base.stickyNote","position":[864,400],"parameters":{"color":7,"width":192,"height":272,"content":"This lets the agent look through the directories to find relevant files"},"typeVersion":1},{"id":"670b4a12-57b3-45dd-8395-bab870d85ef5","name":"Simple Memory","type":"@n8n/n8n-nodes-langchain.memoryBufferWindow","position":[736,384],"parameters":{},"typeVersion":1.3},{"id":"ee6eacbf-8cc6-4600-81d4-1fee1e519ff5","name":"Chat Model","type":"@n8n/n8n-nodes-langchain.lmChatOpenRouter","position":[528,384],"parameters":{"model":"google/gemini-3-flash-preview","options":{"temperature":0.1}},"credentials":{"openRouterApi":{"id":"A5tnKaYbV9QNRMl8","name":"n8n-general-use-mcgarrigle"}},"typeVersion":1},{"id":"84ee2f0c-781b-44f0-a9ae-3c43b079b5f1","name":"List Skills Dirs","type":"n8n-nodes-base.github","onError":"continueRegularOutput","position":[-128,80],"webhookId":"4ec37c56-9985-42e6-baf5-fe4ac627182b","parameters":{"owner":{"__rl":true,"mode":"url","value":"={{ $json.SKILLS_REPOS }}"},"filePath":"/skills","resource":"file","operation":"list","repository":{"__rl":true,"mode":"url","value":"={{ $json.SKILLS_REPOS }}"}},"credentials":{"githubApi":{"id":"aChEhZC5dvKq49tr","name":"liamdmcgarrigle read only"}},"typeVersion":1.1},{"id":"b1568951-5eef-4445-99b5-32a42d973ab2","name":"Remove Skills Dirs & Dot Files","type":"n8n-nodes-base.filter","position":[48,-80],"parameters":{"options":{},"conditions":{"options":{"version":3,"leftValue":"","caseSensitive":true,"typeValidation":"strict"},"combinator":"and","conditions":[{"id":"275c05ea-cde6-4b6a-9916-afe5cda498d9","operator":{"type":"string","operation":"notEquals"},"leftValue":"={{ $json.name }}","rightValue":"skills"},{"id":"50420e00-4d14-44ff-8c86-389fac249fa6","operator":{"type":"string","operation":"notStartsWith"},"leftValue":"={{ $json.name }}","rightValue":"."}]}},"typeVersion":2.3},{"id":"89d669a1-936f-476a-bb4e-bcc7b9bd0782","name":"Remove Errors and Dot Files","type":"n8n-nodes-base.filter","position":[48,80],"parameters":{"options":{},"conditions":{"options":{"version":3,"leftValue":"","caseSensitive":true,"typeValidation":"strict"},"combinator":"and","conditions":[{"id":"275c05ea-cde6-4b6a-9916-afe5cda498d9","operator":{"type":"string","operation":"notExists","singleValue":true},"leftValue":"={{ $json.error }}","rightValue":""},{"id":"b07c5b48-668a-4ca2-9ba2-b84509d8f93d","operator":{"type":"string","operation":"notStartsWith"},"leftValue":"={{ $json.name }}","rightValue":"."}]}},"typeVersion":2.3},{"id":"af4f0351-1c12-4c93-85e7-652910468302","name":"Sticky Note","type":"n8n-nodes-base.stickyNote","position":[-1024,-176],"parameters":{"color":7,"height":352,"content":"## Interact with Chat\nYou can use this with the chat window on the bottom left or by enabling the toggle `Make Available in n8n Chat Hub` to true will let you use it in n8n's [chat hub](https://www.youtube.com/watch?v=s_GHteEnHc4)"},"typeVersion":1},{"id":"83b10b2b-42ab-4c04-a5cc-3d93ac73cf9b","name":"Sticky Note3","type":"n8n-nodes-base.stickyNote","position":[-688,-272],"parameters":{"color":3,"width":256,"height":448,"content":"## TODO Add Skills\nYou can use any skill that is publicly available on GitHub.\n\nPreselected is Anthropic's example [Skills Repo](https://github.com/anthropics/skills) and [Knowledge Work Skills](https://github.com/anthropics/knowledge-work-plugins) Repo\n\nIf you add more, ensure to match the same array syntax"},"typeVersion":1},{"id":"f66b72b8-bb22-4f0e-8f11-8abfb86b9406","name":"Sticky Note4","type":"n8n-nodes-base.stickyNote","position":[480,-128],"parameters":{"color":7,"width":384,"height":272,"content":"## Just a Regular Ol Agent\nYou can build up from here! Add any other tool you want or add to the system prompt to fit your needs!"},"typeVersion":1},{"id":"10857d7f-a1d7-4f36-a5da-a957a49dbb01","name":"Sticky Note5","type":"n8n-nodes-base.stickyNote","position":[864,208],"parameters":{"color":7,"width":464,"height":464,"content":"## How the Agent Uses the Skills\nThe agent uses the `List Files by Path Name` as much as needed to explore the file skills repo. The `Get a File From GitHub` returns the full text from the skill markdown files. \n\nThis mimics exactly how skills work in Claude Code or Open Code\n"},"typeVersion":1},{"id":"12526c15-4ed3-4bb9-b134-613b4693850d","name":"List Files by Path Name","type":"n8n-nodes-base.githubTool","position":[928,512],"webhookId":"b309da8c-2f65-4c37-bb15-7c8a88c482ba","parameters":{"owner":{"__rl":true,"mode":"name","value":"={{ $fromAI('repoUsername') }}"},"filePath":"={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Path', \"`/` would be root, `/folder` would list files in the folder named folder\", 'string') }}","resource":"file","operation":"list","repository":{"__rl":true,"mode":"name","value":"={{ $fromAI('repoName') }}"},"descriptionType":"manual","toolDescription":"List the files and directories inside of dirs.\nUse this tool on any DIR to see the files and other directories inside of it"},"credentials":{"githubApi":{"id":"aChEhZC5dvKq49tr","name":"liamdmcgarrigle read only"}},"typeVersion":1.1},{"id":"8570e9b5-ded9-4dbd-9a32-a6051d90eee8","name":"Get a File From GitHub","type":"n8n-nodes-base.httpRequestTool","position":[1168,528],"parameters":{"url":"=https://api.github.com/repos/{{ $fromAI('repoOwnerUsername', ``, 'string') }}/{{ $fromAI('repoName', ``, 'string') }}/contents/{{ $fromAI('File_Path', ``, 'string') }}","options":{"redirect":{"redirect":{}}},"sendHeaders":true,"authentication":"predefinedCredentialType","toolDescription":"Get text content from a file on github from org, repo name, and path to file","headerParameters":{"parameters":[{"name":"Accept","value":"application/vnd.github.raw+json"},{"name":"X-GitHub-Api-Version","value":"2022-11-28"}]},"nodeCredentialType":"githubApi"},"credentials":{"githubApi":{"id":"aChEhZC5dvKq49tr","name":"liamdmcgarrigle read only"}},"typeVersion":4.4},{"id":"fea5fbb5-1c6c-4c39-a4ba-15f1e2ba7914","name":"Sticky Note6","type":"n8n-nodes-base.stickyNote","position":[-352,-272],"parameters":{"color":7,"width":736,"height":528,"content":"## Fetch & Prepare Skills for Agent\nThis:\n1. Gets the directory from the GitHub Repository\n2. Removes duplicates & irrelevant files to save context\n3. Gets nested `/skills` directory if it exists\n4. Merge all data together to prepare for use in the Agent node's system prompt\n"},"typeVersion":1},{"id":"b424a991-955e-4fdb-9427-91e059d50019","name":"Merge Directory Structures","type":"n8n-nodes-base.merge","position":[256,0],"parameters":{},"typeVersion":3.2},{"id":"218097a1-4675-4b6f-b05d-3aee56e79b46","name":"Sticky Note7","type":"n8n-nodes-base.stickyNote","position":[448,192],"parameters":{"width":224,"height":320,"content":"## Replace Me\nIf you want, replace me with the provider and model you prefer!\n\nMaybe pick one out on our [AI Benchmark](https://n8n.io/ai-benchmark)"},"typeVersion":1}],"pinData":{},"connections":{"Split Out":{"main":[[{"node":"List Root Dirs","type":"main","index":0},{"node":"List Skills Dirs","type":"main","index":0}]]},"Chat Model":{"ai_languageModel":[[{"node":"AI Agent","type":"ai_languageModel","index":0}]]},"Simple Memory":{"ai_memory":[[{"node":"AI Agent","type":"ai_memory","index":0}]]},"List Root Dirs":{"main":[[{"node":"Remove Skills Dirs & Dot Files","type":"main","index":0}]]},"List Skills Dirs":{"main":[[{"node":"Remove Errors and Dot Files","type":"main","index":0}]]},"Set GitHub Repo URLs":{"main":[[{"node":"Split Out","type":"main","index":0}]]},"Get a File From GitHub":{"ai_tool":[[{"node":"AI Agent","type":"ai_tool","index":0}]]},"List Files by Path Name":{"ai_tool":[[{"node":"AI Agent","type":"ai_tool","index":0}]]},"Merge Directory Structures":{"main":[[{"node":"AI Agent","type":"main","index":0}]]},"When chat message received":{"main":[[{"node":"Set GitHub Repo URLs","type":"main","index":0}]]},"Remove Errors and Dot Files":{"main":[[{"node":"Merge Directory Structures","type":"main","index":1}]]},"Remove Skills Dirs & Dot Files":{"main":[[{"node":"Merge Directory Structures","type":"main","index":0}]]}}}

How to Import This Workflow

  1. 1Copy the workflow JSON above using the Copy Workflow JSON button.
  2. 2Open your n8n instance and go to Workflows.
  3. 3Click Import from JSON and paste the copied workflow.

Don't have an n8n instance? Start your free trial at n8nautomation.cloud

Related Templates

Auto-create TikTok videos with VEED.io AI avatars, ElevenLabs & GPT-4

Automate the creation and distribution of trending TikTok videos using AI avatars. This workflow connects Telegram, Perplexity, OpenAI, ElevenLabs, VEED.io, and BLOTATO to generate scripts, synthesize voice, create video, and publish across multiple social platforms. Content creators and marketers can rapidly produce engaging short-form video content without manual editing.

35 nodes

CV Screening with OpenAI

Streamline your hiring process by automating the initial screening of CVs with this powerful workflow. It connects directly to OpenAI to analyze resumes, extracting key information and evaluating candidates based on your criteria. This workflow is ideal for recruiters, HR professionals, and hiring managers who need to quickly assess a large volume of applications, saving significant time and effort in the early stages of recruitment. By automating the parsing of PDF documents and leveraging OpenAI's analytical capabilities, you can efficiently identify top candidates, reduce manual review time, and focus on more strategic aspects of the hiring process. This solution drastically cuts down on the hours spent manually reading CVs, allowing for faster shortlisting and improving overall recruitment efficiency.

11 nodes

Create daily historical AI videos with Gemini, fal.ai, Telegram and YouTube

Automate the creation and publishing of daily historical AI videos. This workflow connects Gemini for script generation, fal.ai for video creation, Telegram for approval, and YouTube for publishing. Content creators or educators can use this to consistently deliver engaging historical content without manual video production. It significantly reduces the time and effort involved in daily video creation and distribution.

30 nodes

Ready to automate with n8n?

Get affordable managed n8n hosting with 24/7 support.