2.5k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill openclaw-flowise-skill- _meta.json281 B
- SKILL.md4.9 KB
Overview
This skill lets you interact with Flowise AI workflows via the Flowise REST API. It supports listing chatflows, sending messages (predictions) to flows, managing conversation sessions, streaming responses, and uploading files. Configure server URL, API key, and default flow in your tools settings before use.
How this skill works
It calls Flowise HTTP endpoints to list flows, retrieve flow details, and post prediction requests to a chosen flow ID. Requests can include question text, sessionId for conversation memory, overrideConfig for model parameters, form objects for structured inputs, and file uploads. The skill parses JSON responses and surfaces results, errors, and streaming output when requested.
When to use it
- You want to send user messages to a Flowise bot or agent.
- You need to list available chatflows or inspect flow details.
- You want to maintain multi-turn context using sessionId.
- You need to override model settings (temperature, maxTokens) per request.
- You need to upload files or send structured form data to a flow.
Best practices
- Store Flowise URL, API key, and default Flow ID in your configuration and validate connectivity with /api/v1/ping first.
- Match user intent to a flow using your tools table; if unspecified, use the default Flow ID.
- Use sessionId consistently to preserve conversation memory across requests.
- Enable streaming for long responses to reduce latency and provide incremental output.
- Include required flow-specific variables or form fields as documented in your configuration for that flow.
Example use cases
- List all chatflows to let a user pick a specialized assistant (support, code, docs).
- Send a message to a code-assistant flow with form parameters like script and device for remote execution tasks.
- Query a RAG/document flow with sessionId and namespace variables for knowledge-base lookups.
- Start a conversation and maintain history by sending sessionId on subsequent prediction calls.
- Upload a PDF to a documentation flow to request summarization or extraction.
FAQ
You can send question, streaming, sessionId, overrideConfig, history, uploads, and a form object for structured inputs.
How do I keep conversation context across calls?
Provide the same sessionId on each prediction to let Flowise maintain conversation memory for that user.