Package Information
Documentation
Taodock for n8n
One node for text, image and video generation across leading AI models —
powered by the Taodock unified AI API.
One API key, one bill, no per-provider credential sprawl.
The node reads the model list authorized by your key at runtime, so new models
appear in the dropdown without a package update.
Features
- Three resources in one node —
Text,Image, andVideo, each with its
own operations and options. - Model discovery from your key — the model dropdown is loaded from
GET /v1/models; nothing is hardcoded. - Media-ready output — generated images and videos are emitted as n8n
binary data, ready for Google Drive, S3, Slack, HTTP upload, or the next node. - Async video handled for you — create a task, optionally wait for
completion, and get the finished MP4 as binary in a single execution. - Usable as an AI Agent tool — the node sets
usableAsTool, so it can be
attached to an n8n AI Agent and driven by an LLM. - Per-item error routing — honours Continue On Fail so one failed item
does not kill a batch run.
Requirements
- A self-hosted n8n instance running on Node.js 22 or newer (community nodes
are not installable on n8n Cloud unless the package is a verified node) - A Taodock API key — create one
Install
Install from the n8n UI:
- Open Settings → Community Nodes → Install.
- Enter
n8n-nodes-taodockand confirm. - Reload n8n; Taodock appears in the node panel.
Manual installation for a local n8n:
cd ~/.n8n/custom # create the directory if it does not exist
npm install n8n-nodes-taodock
For Docker deployments, install the package into the custom extensions folder:
docker exec -it <n8n-container> npm install n8n-nodes-taodock \
--prefix /home/node/.n8n/custom
Credentials
- Create a key at https://taodock.com/keys.
- In n8n, add a Taodock API credential and paste the key.
- n8n verifies the credential against
GET https://taodock.com/v1/models.
The key is sent only to https://taodock.com/v1 as a bearer token.
Operations
| Resource | Operation | What it does | Output |
|---|---|---|---|
| Text | Generate | Text or multimodal chat completion with system message, image URL, temperature, top-p, max tokens, reasoning effort, JSON response format | content, model, usage, raw response |
| Image | Generate | Generate or edit images (up to 10 per call) with aspect ratio, size, quality, seed, negative prompt, PNG/JPEG/WebP output | One item per image, file as binary data |
| Video | Create | Start a video task; set Wait for Completion to poll until done and return the MP4, or 0 to return the task immediately | Task JSON, or binary MP4 when waited |
| Video | Get Status | Inspect an existing video task | Task JSON |
| Video | Download | Download a finished video task | Binary MP4 |
Example: summarise incoming text with any model
- Add a Webhook node that receives a
textfield. - Add a Taodock node → Resource
Text, OperationGenerate. - Set Model from the dropdown, put
={{ $json.text }}in Prompt, and add
a system message such asYou are a concise technical editor. - Enable Response Format → JSON Object and pipe the result into your
downstream nodes using{{ $json.content }}.
Example: generate an image and store it
- Taodock → Resource
Image, OperationGenerate, prompt,Size/Aspect Ratio. - Leave Binary Property as
data. - Connect Google Drive → Upload (or S3) and map the binary property
data.
Privacy and data flow
Prompts, media references, model settings, and the API key are sent tohttps://taodock.com/v1 when a workflow executes. Generated media returned as a
temporary URL is downloaded immediately into n8n binary data. The package
contains no analytics and no telemetry. Details:
PRIVACY.md · Taodock privacy policy.
Compatibility
| Component | Version |
|---|---|
| n8n | 1.x, self-hosted |
| Node.js | 22 or newer |
| n8n nodes API | 1 |
Development
npm ci
npm run lint # eslint
npm run typecheck # tsc --noEmit
npm test # builds, then runs node:test suites
npm run build # n8n-node build
npm run dev # local node development environment
Continuous validation runs on every push via the
Validate workflow.
Releasing
Publishing is automated. Pushing a v* tag runs the
Publish workflow, which verifies the tag matchespackage.json, runs lint, typecheck, tests and build, then publishes to npm with
a provenance attestation:
npm version patch --no-git-tag-version
git commit -am "Release v0.1.5"
git tag v0.1.5
git push origin main --tags
The workflow uses a repository-scoped npm granular token while the package's
Trusted Publisher record is being repaired. Releases still run in GitHub Actions
with --provenance, so npm can verify the source repository, commit, and workflow.
The token is scoped only to the publish step and should be removed after OIDC
publishing is verified. To test the pipeline without publishing, run the workflow
manually with Dry run enabled.
Links
More Taodock integrations
| Integration | Repository |
|---|---|
| WordPress AI provider | taodock-api/taodock-wordpress |
| Dify tools & plugins | taodock-api/taodock-dify-tools |
| AstrBot plugin | taodock-api/taodock-astrbot-plugin |
| MCP server | taodock-api/taodock-mcp |
| TypeScript AI SDK | taodock-api/taodock-ai-sdk |
License
MIT © Taodock