Skip to main content

storywrangler-mcp

MCP server that lets AI agents work with the Storywrangler platform. Modeled on the Svelte team's @sveltejs/mcp and vcsi-starter's @the-vcsi/scrolly-mcp: the server holds no content — it fetches documentation from the docs site's llms.txt exports and dataset metadata from the live registry API, so nothing can drift.

Tools

Tool Source Purpose
list-sections docs site /sections.json Discover documentation sections (guides + per-tag API reference)
get-documentation docs site /{slug}/llms.txt Fetch section markdown, with fuzzy matching
list-datasets API /registry/ List registered datasets with queryable dimensions
get-dataset API /registry/{domain}/{dataset_id} Ground-truth metadata: level_order, filter_values, availability
validate-submission local (storywrangler-schemas) Dry-run a DatasetCreate payload: real Pydantic contract, mirrored server guards, conflation lints, on-disk hive layout checks when reachable

Usage

Remote (streamable HTTP) — the backend mounts this same server at /mcp (see backend/app/main.py). No install needed:

{
  "mcpServers": {
    "storywrangler": {
      "type": "http",
      "url": "https://storywrangler.uvm.edu/mcp/"
    }
  }
}

Keep the trailing slash — /mcp works but costs a 307 redirect per request. On the server, set STORYWRANGLER_DOCS_URL/STORYWRANGLER_URL to localhost addresses so tool fetches don't loop through the public proxy. Note: remote clients validate TLS strictly; this endpoint is unusable externally until the uvm.edu certificate mismatch is fixed.

Local (stdio) — works offline and pins a version:

{
  "mcpServers": {
    "storywrangler": {
      "command": "uvx",
      "args": ["storywrangler-mcp"]
    }
  }
}

CLI mode

Every tool also works as a shell command — the fallback for subagents and CI where no MCP connection exists:

uvx storywrangler-mcp list-sections
uvx storywrangler-mcp get-documentation register querying
uvx storywrangler-mcp list-datasets
uvx storywrangler-mcp get-dataset wikimedia ngrams --full
uvx storywrangler-mcp validate-submission payload.json   # '-' reads stdin

validate-submission exits non-zero on blocking errors, so a pipeline can gate its submit step on it (add --no-disk to skip layout checks).

Configuration

Env var Default Meaning
STORYWRANGLER_DOCS_URL https://storywrangler.uvm.edu Docs site base URL
STORYWRANGLER_URL https://storywrangler.uvm.edu API base URL (same var the SDK uses)
STORYWRANGLER_INSECURE unset Set to 1 to skip TLS verification (the uvm.edu deployment currently serves a mismatched certificate — fix pending)

Point both at http://localhost:5173 / http://localhost:8000 for local development.

Metadata

Release files for storywrangler-mcp 0.1.1

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

Source distribution (sdist)

Source distribution for storywrangler-mcp 0.1.1
File Size Uploaded Provenance Actions
storywrangler_mcp-0.1.1.tar.gz 13.4 kB Without provenance Details for storywrangler_mcp-0.1.1.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for storywrangler-mcp 0.1.1
File Interpreter ABI Platform Actions
storywrangler_mcp-0.1.1-py3-none-any.whl Python 3 none any Details for storywrangler_mcp-0.1.1-py3-none-any.whl

Total release size: 26.5 kB

Release files / storywrangler_mcp-0.1.1.tar.gz

Download URL storywrangler_mcp-0.1.1.tar.gz
Size 13.4 kB
Tags Source
SHA-256 checksum
How to use checksums
ee6ff884ecfb2fe2f87f6bc3751205eeb921f3cd973e9854003eaea9af388039
BLAKE2b-256 checksum
How to use checksums
83fbf44c95a878fec061750ab4b0151a766731f414ca0d70d7c025643ee689ff
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.8.17

Release files / storywrangler_mcp-0.1.1-py3-none-any.whl

Download URL storywrangler_mcp-0.1.1-py3-none-any.whl
Size 13.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
0fe60e9a82074612636d0add0cc635f98ef51d01c1d20dc975e45b2972ebf69a
BLAKE2b-256 checksum
How to use checksums
fc7fa911a063a34fea74dd72c4abda14a5d9f3bdf548998e1bfa82a508322285
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.8.17

Security for storywrangler-mcp 0.1.1

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.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.1 This release

2 release files No provenance

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