Skip to main content

gnosys-strata

Enhanced fork of Klavis Strata with JIT server management and catalog mode.

What's Different

The original Strata has serious scalability issues:

  • Memory hog: Every MCP stays loaded in memory (2.7% each)
  • No caching: Rebuilds tool index on every startup
  • Sync loading: Couldn't even load 5 MCPs without OOM

gnosys-strata fixes this:

Feature Original gnosys-strata
Server loading All at once, sync JIT, async
Memory usage O(n) always O(active)
Tool catalog Rebuilt every time Disk-cached
20+ MCPs OOM crash Works fine
MCP Sets No Yes

New Features

Catalog Mode

Tool schemas are cached to disk. Search across ALL configured servers without loading them:

# Search tools across offline servers
strata search "github issues"

JIT Server Management

Servers only load when you actually use them:

# Via tools
manage_servers(action="connect", server_name="github")  # Load on demand
manage_servers(action="disconnect", server_name="github")  # Free memory
manage_servers(action="list")  # See what's active

MCP Sets

Group servers into named sets, swap between them:

# Coming soon
activate_set("development")  # github, filesystem, git
activate_set("production")   # monitoring, alerting, deploy

Installation

pip install gnosys-strata

Configuration

Same as original Strata - config lives at ~/.config/strata/servers.json:

{
  "mcp": {
    "servers": {
      "github": {
        "command": "npx",
        "args": ["-y", "@modelcontextprotocol/server-github"],
        "env": {"GITHUB_TOKEN": "..."},
        "enabled": true
      }
    }
  }
}

Usage

# Run as MCP server (stdio mode)
strata

# Run as HTTP server
strata run --port 8080

Tools

Core tools (always available):

  • discover_server_actions - Find tools across servers
  • get_action_details - Get tool schema
  • execute_action - Run a tool
  • search_documentation - Search tool docs

New in gnosys-strata:

  • manage_servers - Connect/disconnect servers on demand
  • search_mcp_catalog - Search cached tool index (offline)

License

Apache-2.0 (same as original)

Credits

Fork of Klavis AI Strata. Enhanced by Isaac Wostrel-Rubin.

Metadata

Release files for gnosys-strata 1.1.6

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for gnosys-strata 1.1.6
File Size Uploaded Provenance
gnosys_strata-1.1.6.tar.gz 35.8 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for gnosys-strata 1.1.6
File Interpreter ABI Platform
gnosys_strata-1.1.6-py3-none-any.whl Python 3 none any Details

Total release size: 83.0 kB

Release files / gnosys_strata-1.1.6.tar.gz

Download URL gnosys_strata-1.1.6.tar.gz
Size 35.8 kB
Tags Source
SHA-256 checksum
How to use checksums
8d3d7e42aa08a15d29d73770b35d91d5d714ece253f8241b22336f862e149d0e
BLAKE2b-256 checksum
How to use checksums
8815128ff9157a2fb3f30633269858bff415de09a3517533e22f70483810b0ef
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.6

Release files / gnosys_strata-1.1.6-py3-none-any.whl

Download URL gnosys_strata-1.1.6-py3-none-any.whl
Size 47.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
10ea3d497a750dd3aa992ef0bdfbdae9777772272b45a1ac6cc9e25745806e1f
BLAKE2b-256 checksum
How to use checksums
53aa51f861484740acf89d5073f2652cc804ee4db08335359c446d5714ab26d7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.6

Security for gnosys-strata 1.1.6

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page