IterFact
Paste your data. Get a hosted interactive page at a permanent URL. Share it with anyone.
https://iterfact.com/Opens ChatGPT on the web or desktop and asks it to use the WebMCP tools available here.
Connect straight to this server’s public endpoint.
https://mcp.iterfact.com/mcpWe add this server to your workspace, then open Studio — saved access, one connection to many servers, with a history of what ran.
Last probed Sep 14, 2026 · mcp.iterfact.com
36tools discovered
Showing 25 of 36 from the live probe.
iterFact Create
Create OR upload a hosted iterFact artifact from one public entry point. Use this whenever the user wants to upload, save, store, host, or note something in their iterFact library, or to make, build, render, publish, or turn something into an iterFact. For a plain-text note or paste, pass text=<the text> (plus an optional title) — iterFact wraps and hosts it with brand chrome applied server-side; do NOT hand-author HTML for plain text. Call this for: 'upload this to my library', 'save this note'
iterFact Docs
Return concise iterFact MCP usage guides for Claude and other MCP clients. Use this when tool choice, artifact publishing state, Florida direct-answer behavior, async creation, or URL reflection is unclear. Topics include overview, tool-selection, async-artifact-workflow, template-input-modes, florida-data-filtering, page-publishing-flow, artifact-reflection, no-reinstall-recovery, errors-and-retries, and engine-catalog. WHEN NOT TO USE: do not use this as a substitute for live data or artifact
Authenticate iterFact
Authenticate iterFact in Claude or confirm the current iterFact MCP connection. Use this when the user says authenticate, sign in, connect iterFact, reconnect iterFact, or asks if they are logged in. If the session is not connected, calling this should trigger the iterFact OAuth browser window.
Get iterFact Connection Status
Check whether the current iterFact MCP connection is authenticated, read-only, stale, or needs reconnecting. Use this when uploads, generation, edits, or even status polling start failing and you need to separate auth drift from actual iterFact runtime issues.
Get Account Status
Return the connected iterFact account's current credit balance, billing tier, reload settings, and recent wallet activity. Use this whenever the user asks how many credits they have left, whether their iterFact account is synced, or how recent generation charges affected their balance.
Create Test Artifact
Create a minimal hosted smoke-test artifact in the authenticated user's private Library. This skips planner/model generation and is intended for cheap connector verification before or after deploys. WHEN NOT TO USE: do not use this for real user content, production pages, template demos, or anything that should consume planner/model quality review.
Run Mcp Smoke Test
Run the production iterFact MCP smoke-test sequence from one tool call: public tool catalog, auth/profile, minimal private test artifact creation, server-side artifact serving verification, and Florida direct-answer router check. Use this when a client cannot reliably chain create_test_artifact, verify_artifact_serves, and recommend_iterfact_products itself.
Create Artifact Draft
Step 1 for longform server-generated builds from source material: create a draft scaffold that can be filled section by section before publication. Use this only when iterFact is assembling the document on the server. If Claude already knows the story, wrapper, and planned sections, pass artifact_plan to skip server-side scaffold planning. WHEN NOT TO USE: Do not use this for already-authored final HTML, cheap smoke tests, ordinary one-shot reports, or existing artifact edits.
Append Section
Step 2 for longform chunked builds: append one source-backed section to a server-generated iterFact draft. This belongs to the longform draft workflow, not the finished-HTML hosting path. Pass idempotency_key (a short stable client key) when retrying after a timeout/disconnect: an exact retry returns the original result without appending the section twice. WHEN NOT TO USE: do not call this unless create_artifact_draft has already returned a draft artifact_id.
Finalize Artifact
Step 3 for longform chunked builds: finalize a draft render and kick off publication for a server-generated artifact. This is not the hosting path for already-authored final HTML. By default this continues in the background so long renders do not time out; then poll get_artifact_status for completion. If skip_critic=true, iterFact skips the editorial critic and publishes as soon as delivery validation passes. If your client does not expose get_artifact_status yet, calling finalize_artifact again
Get Artifact Status
Poll the current status of any iterFact artifact - whether it is planning, rendering, published, or failed. TRUST these authoritative fields first: `state` (working|done|failed - the single source of truth), `public_url` (the live URL to share), and `current_version`/`current_version_label` (which version is live now); `published_pages` lists the named @handle/slug pages serving it. A done artifact never reports Waiting/0%; a mid-flight revision adds revision_in_progress, and a failed attempt ad
Verify Artifact Serves
Server-side smoke check for a returned iterFact artifact URL. Pass artifact_id or a returned iterfact.com/a.iterfact.com URL; the tool performs the fetch from the MCP server side and returns status, content type, and content length. Pass expect_text to additionally assert a substring is present in the served page's visible text - the way to PROVE an edit is live when your sandbox cannot fetch the URL itself. WHEN NOT TO USE: do not use this to create, edit, or publish artifacts; it only proves w
Send To Library
Repair / re-sync step: ensure an EXISTING hosted iterFact artifact is saved in the authenticated user's private Library. Requires an existing artifact_id. Use this only when a client is unsure whether an already-created artifact was synced. WHEN NOT TO USE: do not use this to upload or create NEW content — use iterfact_create (with text= for a plain-text note, or a prompt) instead. Do not call this before an artifact exists.
Recommend iterFact Products
Recommend the best iterFact templates and Atlas tools for a user's goal before rendering anything. Use this first for questions like 'does iterFact have Florida real estate data?', 'what Florida property data is available?', 'what company data does iterFact have?', or 'what can iterFact do for X?' so the user sees the real product menu before any brainstorming. Default payload_mode is minimal so Claude gets the direct answer and top products without static taxonomy bloat; pass payload_mode='full
List iterFact Products
List callable iterFact products across templates and Atlas tools. Use this when the user asks what iterFact can do, asks whether iterFact has Florida real estate/property data or company data, or wants the full callable catalog before choosing a product. WHEN NOT TO USE: do not use this to answer a specific user goal when recommend_iterfact_products can rank the best options and return direct data previews.
List Templates
List vault-backed iterFact templates, including fixed primary templates and composable panels that keep structure frozen while only data updates. Use the returned supported_input_modes, preferred_input_mode, input_guidance, and quick_example to tell the user how they can add data before rendering anything. This is a metadata-only list view; call get_template_contract(template_id) for a template's full input schema. WHEN NOT TO USE: do not use this as the broad product menu; use recommend_iterfac
Match Template
Recommend the best matching fixed iterFact templates for a prompt before generating anything. Before you render, tell the user the matched template's supported_input_modes, preferred_input_mode, and quick_example when one is available. WHEN NOT TO USE: do not call this to create or publish the artifact; after choosing a template, call iterfact_create with template_id.
List Datasets
List the structured datasets iterFact can query through the generic data interface (for example Florida property data). Each entry gives the dataset id and its available actions. Then call describe_dataset(dataset) for an action's parameters and query_dataset(dataset, action, params) to run it. Admin-only actions are omitted for non-admin callers.
Describe Dataset
Describe one iterFact dataset: its actions, each action's allowed parameters, and source attribution. Call this before query_dataset to shape valid input on the first try. Admin-only actions are omitted for non-admin callers.
Query Dataset
Run a declared action on an iterFact dataset, e.g. query_dataset('florida_property', 'lookup', {'county_code': '58', 'parcel_id': '...'}). This is the single dataset-agnostic data verb: new data domains are added as catalog/registry entries, not as new tools. Use list_datasets and describe_dataset first. Unknown datasets/actions, unsupported parameters, and admin-only actions for non-admin callers are refused.
Apply Artifact Feedback
Revise an existing published iterFact in place from saved feedback notes, a prompt-only instruction, or both instead of starting over. By default this runs in the background and republishes under the same artifact id. Non-admin MCP callers can revise template-backed and engine-composed artifacts, but raw uploaded HTML revisions remain admin-only. WHEN NOT TO USE: do not use this for simple one-instruction edits when edit_artifact is enough, or for creating/publishing a new artifact.
Edit Artifact
Edit an existing iterFact artifact by applying a natural language instruction. The artifact is revised in place under the same URL - the user's library version history is preserved. Use this when a user says 'edit my revenue dashboard' or 'make the charts bigger on that report'. For admin-owned upload-bridge artifacts you can also use edit_mode='find_replace' with literal find and replace strings for surgical HTML edits. Hosted upload-bridge artifacts keep iterFact's serve-time artifact chrome a
Get Artifact Data
Return the stored deterministic data payload for one of the signed-in user's template-backed or engine-composed artifacts, including normalized_data or sections data, input metadata, and revision count.
Get Artifact Feedback
Read feedback notes saved against an existing iterFact artifact, including notes left from the published Iterate link.
Get Artifact Sections
List the saved plan sections for an existing iterFact artifact so callers can target a single section for regeneration.
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Directory coverage for brandsPaste your data. Get a hosted interactive page at a permanent URL. Share it with anyone.
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
Operate IterFact? Verify ownership to take over this directory entry.
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.
MCPBundles probed 36 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
No provider sign-in was required during MCPBundles' probe. Your client may still need MCPBundles credentials depending on how you connect.
MCPBundles is an independent platform built on the open Model Context Protocol standard. Not affiliated with Anthropic PBC or Claude.