Skip to content
Agentmart
Submit a tool

ai-services

Use this skill for creating AI agents, setting up knowledge bases (RAG), configuring LLM models, building tools (API/MCP), managing conversations, streaming…

Skill by rezwanx

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/rezwanx-ai-services

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use this skill for creating AI agents, setting up knowledge bases (RAG), configuring LLM models, building tools (API/MCP), managing conversations, streaming chat, or querying LLMs directly on SELISE Blocks. Note: this API uses snake_case throughout.

Use cases: backend, ml, productivity, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/ai-services

What's inside

Skill name
ai-services
Description
Use this skill for creating AI agents, setting up knowledge bases (RAG), configuring LLM models, building tools (API/MCP), managing conversations, streaming chat, or querying LLMs directly on SELISE Blocks. Note: this API uses snake_case throughout.
License
MIT
Instructions
153 lines in SKILL.md · view on GitHub

Files (42)

  • SKILL.md
  • change-agent-status.md · actions/
  • chat-agent.md · actions/
  • chat-sse.md · actions/
  • create-agent.md · actions/
  • create-api-tool.md · actions/
  • create-kb-folder.md · actions/
  • create-mcp-tool.md · actions/
  • create-model.md · actions/
  • delete-agent.md · actions/
  • delete-conversation.md · actions/
  • delete-kb.md · actions/
  • delete-tool.md · actions/
  • get-agent.md · actions/
  • get-agents.md · actions/
  • get-conversations.md · actions/
  • get-model.md · actions/
  • get-models.md · actions/
  • get-provider-models.md · actions/
  • get-tools.md · actions/
  • ingest-kb-link.md · actions/
  • ingest-kb-qa.md · actions/
  • ingest-kb-text.md · actions/
  • initiate-conversation.md · actions/
  • publish-agent.md · actions/
  • query-lmt.md · actions/
  • stream-query-lmt.md · actions/
  • test-kb-retrieval.md · actions/
  • test-tool-action.md · actions/
  • update-agent-ai-config.md · actions/
  • update-agent-persona.md · actions/
  • upload-kb-file.md · actions/
  • validate-model.md · actions/
  • contracts.md
  • chat-flow.md · flows/
  • create-agent-flow.md · flows/
  • manage-models.md · flows/
  • query-lmt-flow.md · flows/
  • setup-knowledge-base.md · flows/
  • frontend-blazor-tailwind.md
  • frontend-blazor.md
  • frontend-react.md

SKILL.md (first lines)

# AI Services Skill

## Purpose

Handles all AI agent management, knowledge base ingestion, tool configuration, model setup, and conversation operations for SELISE Blocks via the blocksai-api v1 API.

> **Important:** This API uses **snake_case** (Python/FastAPI style) throughout. All request and response fields use `snake_case` — e.g., `project_key`, `agent_id`, `is_success`, `kb_ids`, `tool_ids`. Do NOT use camelCase when constructing requests or reading responses.

Must run get-token before any action in a session.

---

## When to Use

Example prompts that should route here:
- "Create an AI customer support agent with a knowledge base"
- "Upload product docs as a knowledge base for RAG"
- "Configure GPT-4 as the model for my agent"
- "Build a chat interface that streams responses from an agent"
- "Set up an MCP tool so the agent can call external APIs"

---

## Execution Context

Before executing any action or flow from this skill, read `../core/execution-context.md` for the required supporting files, load order, and cross-domain orchestration rules.

---

## Base Path

All endpoints are prefixed with: `$API_BASE_URL/blocksai-api/v1`

---

## Intent Mapping

Use this table to route user requests. Check `flows/` first — if a flow covers the request, use it. For single-action requests, go directly to the action.

| User wants to... | Use |
|------------------|-----|
| Create an AI agent from scratch | `flows/create-agent-flow.md` |
| Send a prompt to an LLM directly without an agent | `flows/query-lmt-flow.md` |
| Stream a response from a language model | `flows/query-lmt-flow.md` |
| Configure an agent's model, KB, tools | `flows/create-agent-flow.md` |
| Publish an agent to marketplace | `actions/publish-agent.md` |
| Set up a knowledge base with files/text/links | `flows/setup-knowledge-base.md` |
| Upload files to a knowledge base | `actions/upload-kb-file.md` |
| Add text content to a knowledge base | `actions/ingest-kb-text.md` |
| Add Q&A pairs to a knowledge base | `actions/ingest-kb-qa.md` |
| Crawl a URL into a knowledge base | `actions/ingest-kb-link.md` |
| Test knowledge base retrieval | `actions/test-kb-retrieval.md` |
| Chat with an AI agent | `flows/chat-flow.md` |
| Start a conversation session | `actions/initiate-conversation.md` |
| Send a message to an agent | `actions/chat-agent.md` |
| Stream a response from an agent | `actions/chat-sse.md` |
| Query the LLM directly (no agent) | `actions/query-lmt.md` |
| Stream a direct LLM query | `actions/stream-query-lmt.md` |
| Add an AI model config | `flows/manage-models.md` |
| List available models | `actions/get-models.md` |
| Validate a model's API key | `actions/validate-model.md` |
| Create an API tool for an agent | `actions/create-api-tool.md` |
| Create an MCP server tool | `actions/create-mcp-tool.md` |
| Test a tool action | `actions/test-tool-action.md` |
| List all tools | `actions/get-tools.md` |
| Delete a tool | `actions/delete-tool.md` |
| List agents | `actions/get-agents.md` |
| Get agent details | `actions/get-agent.md` |
| Update agent name/description/persona | `actions/update-agent-persona.md` |
| Update agent AI configuration | `actions/update-agent-ai-config.md` |
| Enable or disable an agent | `actions/change-agent-status.md` |
| Delete an agent | `actions/delete-agent.md` |
| List conversation sessions | `actions/get-conversations.md` |
| Delete a conversation session | `actions/delete-conversation.md` |

---

## Flows

| Flow | File | Description |
|------|------|-------------|
| create-agent-flow | flows/create-agent-flow.md | Create agent → configure AI settings → add KB → add tools → publish |
| setup-knowledge-base | flows/setup-knowledge-base.md | Create folder → upload files/text/links → test retrieval → attach to agent |
| chat-flow | flows/chat-flow.md | Initiate conversation → send messages → handle streaming → manage session |
| manage-models | flows/manage-models.md | List providers → get available models → create model config → validate API key |

---

## Action Index

### Agents

| Action | File | Description |
|--------|------|-------------|
| create-agent | actions/create-agent.md | Create a new AI agent from a description |
| update-agent-persona | actions/update-agent-persona.md | Update agent name, description, and persona |
| update-agent-ai-config | actions/update-agent-ai-config.md | Update model, temperature, system prompt, KB and tool attachments |
| change-agent-status | actions/change-agent-status.md | Enable, disable, or archive an agent |
| delete-agent | actions/delete-agent.md | Permanently delete an agent |
| get-agents | actions/get-agents.md | List agents with filter and pagination |
| get-agent | actions/get-agent.md | Get a single agent's full details |
| publish-agent | actions/publish-agent.md | Publish an agent to the marketplace |

### Knowledge Base

| Action | File | Description |
|--------|------|-------------|
| upload-kb-file | actions/upload-kb-file.md | Upload and index a file (PDF, DOCX, TXT, etc.) |
| ingest-kb-text | actions/ingest-kb-text.md | Ingest raw text content into a KB folder |
| ingest-kb-qa | actions/ingest-kb-qa.md | Ingest Q&A pairs into a KB folder |
| ingest-kb-link | actions/ingest-kb-link.md | Crawl and index a URL into a KB folder |
| create-kb-folder | actions/create-kb-folder.md | Create a KB folder with embedding configuration |
| delete-kb | actions/delete-kb.md | Delete a knowledge base entry |
| test-kb-retrieval | actions/test-kb-retrieval.md | Test retrieval quality with a query against an agent |

### Tools

| Action | File | Description |
|--------|------|-------------|

From rezwanx/blocks-ai-skills, MIT.

More from rezwanx/blocks-ai-skills

Signals

Stars
0
Forks
0
Open issues
1
Last commit (this folder)
Mar 27, 2026 (7 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from rezwanx

Cite this page

ai-services skill. Agent Mart. https://agentmart.dev/skills/rezwanx-ai-services (last updated 2026-10-24)