Have an AI agent? Let it explore Baltor for you. Any assistant works, personal or coding.

Connect your agent
What your agent reads

The request it copies

Explore baltor.ai for me and report back. Start with https://baltor.ai/llms.txt, then read https://baltor.ai/.well-known/agent-card.json and https://baltor.ai/.well-known/mcp.json. Tell me which plan (Feeds at $4.99/mo, Pro at $29/mo, or Max at $49/mo) fits my work, what you would connect, and the setup steps at https://baltor.ai/connect. Stay read-only: read public pages and agent files, and check with me before any sign-in, purchase or change. If a step needs a key, name it (BALTOR_SERVICE_TOKEN) and I will add it myself.

Discovery files on baltor.ai

Read-only discovery is open. Keep BALTOR_SERVICE_TOKEN in your own environment.

Baltor
Sign in Get started

Start every task with working code.

Each step gets the skills, instructions and reusable code it needs, while you keep your own agent, models and machine.

Turn your internal API into an MCP server, step 2
step-2-build-server/
├── CLAUDE.md
├── .claude/skills/
│ └── openapi-to-mcp-server/
├── .mcp.json
└── .baltor/step.lock.json
# exact versions and checksums
  1. Instructions CLAUDE.mdWhat this step should do and how to check it.
  2. One chosen skill openapi-to-mcp-serverPicked by your agent, at its exact version.
  3. Tools it may call .mcp.jsonMCP server settings for this step.
See how it works

Find the code before writing it again.

Every item carries its source, version, licence and checksum, so your agent picks up the code it needs and builds on it.

Browse the library
MCP tool

Search the whole library with a query in plain words. Each result shows its kind, licence, size and checksum.

Sample call
intelligence_search
query "MCP server from an OpenAPI spec"
# 1 openapi-to-mcp-server, skill, MIT
# 2 mcp-tool-schema-checks, code module

provisioning_read

MCP tool

Download the result your agent chose by its identity, with expected_digest set to that version's SHA-256 checksum.

Sample call
provisioning_read
identity "openapi-to-mcp-server"
expected_digest "3f9c1a07…"
Checksum matches, saved for this step

Keep the large model for the hard part.

Each step can run on the model that fits it, through the model access you already have.

  • Provider key
  • Ollama Cloud (glm-5.3-flash)
  • Ollama Cloud
  • Machine on your network
FeedsProMax

Match the right model to every step

  1. 1Read benchmark, cost and latency sourcesFeeds
  2. 2Check which models fit your hardwareFeeds
  3. 3Pull settings and subagents that route each stepPro
  4. 4Run the job overnight within the call limits you setMax

Your machine stays yours.

See access and data
  • Your project stays local

    Your code and repository stay on your machine. Baltor keeps your account and download information.

  • Your model keys stay with you

    Your agent calls your models through the provider connection you control.

  • You choose what runs

    Your agent runs the files you select, with the permissions you choose.

Built for the agent you run.

Get set up

17 more over MCPhttps://baltor.ai/mcp

Cursor, Gemini CLI, GitHub Copilot, Windsurf, Cline, Roo Code, Kilo Code, Aider, Goose, Amp, Kiro, Zed, Qwen Code, Crush, Factory Droid, Hermes Agent and OpenClaw connect over MCP with one entry in their MCP server settings.

Add the full library.

Search all 1,560,856 files from the agent you already use.

Pro

$29a month

Includes Agent Feeds + Skills and agent files

Max $49 a month Public preview Adds Overnight / AFK Work for overnight and AFK runs.