MCP server · registry snapshot 2026-10-23 · last tested 2026-10-23

ai.zerowidth/workbench

Build, run and publish Workbench flows, tasks, shims and knowledge bases. — the publisher's description

1 of 2 record checks metRead-only · tools listed 2026-10-23Topic: developer-tools · AI-classifiedPricing: freemium · AI-classifiedversion 1.0.0streamable HTTP61 tools

Summary

ai.zerowidth/workbench is a remote MCP server listed in the official MCP registry at version 1.0.0. On 2026-10-23 ProtocolProbe connected to it in read-only mode and read its inventory: 61 tools, 134 resources and 0 prompts.

Record checks

1 of 2 record checks met · 7 do not apply

met: Registry entry is active and says how to run it
not met: Source repository link opens
not applicable: Repository is not archived
not applicable: Code pushed to the repository in the 90 days before our check
not applicable: Repository publishes a security policy (SECURITY.md)
not applicable: Package names its licence with a standard (SPDX) identifier
not applicable: The version the registry lists can be installed and is not withdrawn
not applicable: npm package carries build provenance (attestation)
not applicable: No known critical or high advisory in the package or its dependencies

Not ranked by downloads: a remote server publishes no download figure
Not ranked by stars: no public GitHub repository

From public records, not a safety rating. Each check with its reason · Is it safe to install?

Key facts

Runs: remote (hosted by the publisher)
Transport: streamable HTTP
Sign-in: none needed to open a session
Tools: 61
Described as changing data: 34 of 61 toolsAI-classified
Mainly works on: files and code on the machine it runs onAI-classified
Topic: developer-toolsAI-classified
Pricing: freemiumAI-classified
Last tested: 2026-10-23

Add to your client

Observed fact

This server accepted a session without any credential and lists 26 tools whose name or declaration suggests they change data (workbench_flows_update, workbench_flows_publish, workbench_flows_delete, workbench_flows_schedules_update, workbench_flows_schedules_delete and 21 more). That is our reading of the names; we did not call them. Source: our connection, 2026-10-23.

Worth knowing before you connect it

34 tools are described as changing something (workbench_flows_update, workbench_flows_publish, workbench_flows_delete, …): our AI model's reading of the descriptions, not something we tested, and the server opened a session without any credential. We did not call them.

Each line comes from a section below, with its source and date.

How fresh this is

Live-tested: yes, a real session (not a text-only review)
Last tested: 2026-10-23
Mode: read-only (no tool calls)
Latest version: 1.0.0, published 2026-10-07 (registry, as of the 2026-10-23 snapshot)
First listed version: 2026-10-07 (earliest version still in the registry; older removed versions would be earlier)
Registry data: snapshot 2026-10-23
Package and repository: checked 2026-10-22
Engine: mcp-runtime-engine@0.1.0 · sdk@1.32.0

Changes since the last refresh

New in the registry: not in the list of 2026-10-04, listed on 2026-10-23 with version 1.0.0 (published 2026-10-07).

Source: the official MCP Registry list on 2026-10-04 and on 2026-10-23.

What it is

What does this server offer?

Observed inventory

61tools

134resources

0prompts

What it mainly works on, as our AI model read its tool descriptions: files and code on the machine it runs on.

34 of 61 tools are described as changing something (creating, editing, deleting, sending or running).

search_docson 140 other serversreads or searches
“Search ZeroWidth product documentation.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- querystringrequired“Search query — keywords or natural-language phrase.”
- limitinteger“Max number of results.”
get_docon 42 other serversreads or searches
“Fetch the full Markdown body of a specific docs page by its slug.”
Publisher declares: read-only yes · destructive no · open-world no
1 input, 1 required
- slugstringrequired“Page slug.”
list_docson 31 other serversreads or searches
“Enumerate all available docs pages, optionally filtered by product (e.g.”
Publisher declares: read-only yes · destructive no · open-world no
1 input
- productstring“Optional product slug filter (e.g.”
workbench_flows_revisions_listreads or searches
“The flow's revision history, newest first: published versions carry a `version` label and note; entries with version null are draft autosaves.”
Publisher declares: read-only yes · destructive no · open-world no
4 inputs, 1 required
- flowIdstringrequired“Flow id.”
- limitinteger“Entries to return, default 20.”
- publishedOnlyboolean“true = only labeled published versions (default false).”
- workspacestring“Workspace slug.”
workbench_flows_updatecreates or changes datadescribed as changing data
“Changes a flow's name, description, visibility, tags, or archived state — the metadata around the flow, NOT its orchestration (prompts and nodes change through workbench_flows_edit_text).”
Publisher declares: read-only no · destructive yes · open-world no
8 inputs, 1 required
- flowIdstringrequired“Flow id, from workbench_flows_list.”
- flowNamestring“New name.”
- descriptionstring or null“New description; null clears it.”
- visibilitystring“Who can see it: PRIVATE (only the user), WORKSPACE (every member, the default), or SHARED (specific people, granted afterwards).”
- tagsarray“Replacement tag set (lowercase labels).”
- archivedboolean“true archives, false restores.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_publishcreates or changes datadescribed as changing data
“Snapshots the flow's current DRAFT as a named published version — the version schedules (workbench_flows_schedule), guest share links (workbench_flows_share_create), public-API runs, and 'published'-…”
Publisher declares: read-only no · destructive yes · open-world yes
6 inputs, 2 required
- flowIdstringrequired“Flow to publish.”
- versionstringrequired“Version label, e.g.”
- notestring“What changed — a commit message.”
- untriedboolean“true only when the person asked to publish this draft without trying it first.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_deletedeletesdescribed as changing data
“Deletes a flow.”
Publisher declares: read-only no · destructive yes · open-world no
3 inputs, 1 required
- flowIdstringrequired“Flow id, from workbench_flows_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_forkcreates or changes datadescribed as changing data
“Copies a PUBLIC flow from another workspace (a template) into this one as a new draft the user owns.”
Publisher declares: read-only no · destructive no · open-world no
4 inputs, 1 required
- flowUuidstringrequired“The source flow's public uuid.”
- forkedFromVersionstring“Published version label to copy.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_schedules_listreads or searches
“Every schedule on one flow: cadence, whether it's paused (`enabled: false`), the next fire time, delivery, and the last run's status.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- flowIdstringrequired“Flow id, from workbench_flows_list.”
- workspacestring“Workspace slug.”
workbench_flows_schedules_updatecreates or changes datadescribed as changing data
“Changes one schedule: `enabled: false` pauses it (configuration kept), `enabled: true` resumes and recomputes the next fire time from now; `interval` / `scheduleTime` / `scheduleTimezone` / `schedule…”
Publisher declares: read-only no · destructive yes · open-world no
13 inputs, 2 required
- flowIdstringrequired“Flow the schedule belongs to.”
- scheduleIdstringrequired“Schedule id, from workbench_flows_schedules_list.”
- labelstring or null
- scheduleTimestring or null
- scheduleTimezonestring or null
- scheduleDayOfWeekinteger or null
- recurrenceobject or null“When it runs, like a calendar event repeats; use it for anything beyond a plain hourly/daily/weekly at one time (then omit interval).”
- deliverystring
- intervalstring“New cadence: hourly, daily, or weekly.”
- inputobject“New fixed input envelope — {"kind":"chat","messages":[…]} or {"kind":"form","values":{…}}.”
- enabledboolean“false pauses, true resumes.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_schedules_deletedeletesdescribed as changing data
“Removes a schedule for good.”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- flowIdstringrequired“Flow the schedule belongs to.”
- scheduleIdstringrequired“Schedule id, from workbench_flows_schedules_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_schedules_run_nowruns code or commandsdescribed as changing data
“Runs one iteration of a schedule immediately through the real scheduled pipeline (same input, same delivery — the email or channel it normally posts to) without moving its cadence.”
Publisher declares: read-only no · destructive no · open-world yes
4 inputs, 2 required
- flowIdstringrequired“Flow the schedule belongs to.”
- scheduleIdstringrequired“Schedule id, from workbench_flows_schedules_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”

Show 49 more tools

workbench_flows_shares_listreads or searches
“Every guest link on one flow, newest first: label, whether it's still live, the shareUrl (live links only — dead ones have no URL), expiry, and how many guests and messages it has seen.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- flowIdstringrequired“Flow id, from workbench_flows_list.”
- workspacestring“Workspace slug.”
workbench_flows_shares_revokedeletesdescribed as changing data
“Kills one guest link immediately — anyone holding it gets a closed page from then on.”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- flowIdstringrequired“Flow the link belongs to.”
- shareIdstringrequired“Share id, from workbench_flows_shares_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_scenarios_listreads or searches
“A flow's variables (each {{token}} in an agent's prompt, with its type) and every saved scenario with its values.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- flowIdstringrequired“Flow id, from workbench_flows_list.”
- workspacestring“Workspace slug.”
workbench_flows_scenarios_createcreates or changes datadescribed as changing data
“Saves a named scenario: values for the flow's variables, e.g.”
Publisher declares: read-only no · destructive no · open-world no
5 inputs, 2 required
- flowIdstringrequired“Flow id.”
- scenarioNamestringrequired“What it's called in the menu, e.g.”
- valuesobject“Values by variable name, e.g.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_scenarios_updatecreates or changes datadescribed as changing data
“Renames a scenario or changes its values.”
Publisher declares: read-only no · destructive yes · open-world no
6 inputs, 2 required
- flowIdstringrequired“Flow id.”
- scenarioIdstringrequired“Scenario id, from workbench_flows_scenarios_list.”
- scenarioNamestring“New name.”
- valuesobject“Values by variable name, e.g.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_scenarios_deletedeletesdescribed as changing data
“Deletes one scenario.”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- flowIdstringrequired“Flow id.”
- scenarioIdstringrequired“Scenario id, from workbench_flows_scenarios_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_attach_knowledge_basecreates or changes datadescribed as changing data
“Attaches a knowledge base to an AGENT flow's draft, the same way picking it in the agent editor does: a docs knowledge base becomes a search tool the model calls when a question needs it; a tabular o…”
Publisher declares: read-only no · destructive no · open-world no
4 inputs, 2 required
- flowIdstringrequired“The agent flow, from workbench_flows_scaffold or workbench_flows_list.”
- knowledgeBaseIdstringrequired“The knowledge base id, from workbench_kb_create or workbench_kb_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_runruns code or commandsdescribed as changing data
“Execute a flow and return its outputs synchronously.”
Publisher declares: read-only no · destructive no · open-world yes
7 inputs, 1 required
- approvalIdstring“Re-call with the approvalId from a needs_confirmation response after the user approves.”
- flowIdstringrequired“Flow id to run.”
- inputobject“Flow input envelope (RunFlowInput).”
- sourcestring“Which version to run.”
- versionstring“Published version label (only with source:"published").”
- workspacestring“Workspace slug.”
- timeoutMsinteger“Max runtime in ms (also the hard cost ceiling).”
workbench_flows_schedulecreates or changes datadescribed as changing data
“Set a specific built flow to run automatically on a cadence — the deterministic counterpart to a routine.”
Publisher declares: read-only no · destructive no · open-world no
11 inputs, 2 required
- flowIdstringrequired“Flow to schedule (id from workbench_flows_list).”
- labelstring or null
- scheduleTimestring or null
- scheduleTimezonestring or null
- scheduleDayOfWeekinteger or null
- recurrenceobject or null“When it runs, like a calendar event repeats; use it for anything beyond a plain hourly/daily/weekly at one time (then omit interval).”
- deliverystring
- intervalstring“Plain cadence: hourly, daily, or weekly.”
- inputobjectrequired“Fixed input envelope every run executes with — {"kind":"chat","messages":[…]} or {"kind":"form","values":{…}}, the same shape workbench_flows_run takes.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_listreads or searches
“Lists every flow in the active workspace the caller can see.”
Publisher declares: read-only yes · destructive no · open-world no
1 input
- workspacestring“Workspace slug.”
workbench_executions_getreads or searches
“The flow's stack traces: recent executions with per-node timelines — what each node received, produced, how long it took, and the exact error when one failed.”
Publisher declares: read-only yes · destructive no · open-world no
4 inputs, 1 required
- flowIdstringrequired“Workbench flow id.”
- executionIdstring“Specific execution to inspect.”
- limitinteger“How many recent executions (default 3, max 10).”
- workspacestring“Workspace slug.”
workbench_flows_getreads or searches
“One flow.”
Publisher declares: read-only yes · destructive no · open-world no
4 inputs, 1 required
- flowIdstringrequired“Flow id.”
- viewstring“`summary` (default): the flow's fields plus one line per node (id, type, label, and how long its prompt is) and the links — enough to pick a node.”
- nodeIdstring“With view `node`: which node.”
- workspacestring“Workspace slug.”
workbench_flow_authoring_guidereads or searches
“READ THIS FIRST before writing or editing any raw flow orchestration JSON.”
Publisher declares: read-only yes · destructive no · open-world no
Takes no inputs.
workbench_node_catalog_getreads or searches
“The ground truth for what nodes exist and what their ports actually are — verify against this instead of recalling.”
Publisher declares: read-only yes · destructive no · open-world no
7 inputs
- workspacestring“Workspace slug.”
- flowIdstring“A flow whose pinned imports should appear as `imported-<uuid>` nodes.”
- slugstring“Exact node slug → full detail for that one node (ports + settings).”
- qstring“Keyword over slug / name / tagline / description.”
- categorystring“Filter by category.”
- limitinteger“Results per page, 1-50.”
- offsetinteger“Pagination offset.”
workbench_flows_scaffoldcreates or changes datadescribed as changing data
“Creates a runnable first-draft flow from a spec you author: pick the simplest pattern that fits (classifier for read-and-bucket, structurer for transform/extract/draft-for-review, agent for genuinely…”
Publisher declares: read-only no · destructive no · open-world no
13 inputs, 2 required
- workspacestring“Workspace slug.”
- opportunityIdstring“Compass change (from compass_opportunities_list) this draft implements.”
- modestringrequired“Pattern editor the draft opens in.”
- flowNamestringrequired“Imperative + specific, ≤80 chars.”
- descriptionstring“1-2 sentences: what it does, where it fits.”
- systemPromptstring“The draft's behavior — production-quality first attempt with [STUB: ...] / [FILL IN: ...] markers where reality is missing.”
- modelstring“The model node slug the draft runs on, from workbench_node_catalog_get, that the workspace allows.”
- inputDescriptionstring“What the flow's single input carries.”
- responseSchemaJsonstring“structurer only: the output JSON Schema, JSON-encoded as a string (object root, required fields, additionalProperties false).”
- categoriesarray“classifier only: 2-12 buckets.”
- stubNotesstring“What's stubbed / missing / worth wiring next.”
- variablesarray“agent / interviewer: types and descriptions for the {{tokens}} in systemPrompt.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_flows_edit_textcreates or changes datadescribed as changing data
“Applies ONE precise text replacement to a flow's DRAFT orchestration — Edit-tool semantics: `find` must be the EXACT current text, copied character-for-character from workbench_flows_get, and must oc…”
Publisher declares: read-only no · destructive yes · open-world no
6 inputs, 4 required
- workspacestring“Workspace slug.”
- flowIdstringrequired“Flow to edit.”
- findstringrequired“EXACT current text (≥3 chars), copied verbatim — never reconstructed from memory.”
- replacestringrequired“Replacement text.”
- notestringrequired“Why this edit.”
- approvalIdstring“Approval id from a prior needs_confirmation envelope.”
workbench_flows_share_createcreates or changes datadescribed as changing data
“Creates a no-account guest link where anyone can chat with the flow's latest PUBLISHED revision at workbench's /s/<token> page — the fastest way to put a working flow in a stakeholder's hands ('here,…”
Publisher declares: read-only no · destructive no · open-world yes
5 inputs, 1 required
- workspacestring“Workspace slug.”
- flowIdstringrequired“Flow to share.”
- labelstring“Who/what this link is for, e.g.”
- expiresInDaysinteger“Days until the link dies.”
- approvalIdstring“Approval id from a prior needs_confirmation envelope.”
workbench_kb_getreads or searches
“One knowledge base: name, recipe (docs / tabular / graph), status, document and chunk counts, embedding model, and ingestion settings.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- kbIdstringrequired“Knowledge base id.”
- workspacestring“Workspace slug.”
workbench_kb_documents_listreads or searches
“Every document ingested into the KB's current content: id, name, type, size, chunk count.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- kbIdstringrequired“Knowledge base id, from workbench_kb_list.”
- workspacestring“Workspace slug.”
workbench_kb_ingestion_statusreads or searches
“The state of one ingestion run started by workbench_kb_add_documents: PENDING / RUNNING / DONE / FAILED, chunks written, and the error when it failed.”
Publisher declares: read-only yes · destructive no · open-world no
3 inputs, 2 required
- kbIdstringrequired“Knowledge base id.”
- runIdstringrequired“Ingestion run id, from workbench_kb_add_documents.”
- workspacestring“Workspace slug.”
workbench_kb_deletedeletesdescribed as changing data
“Deletes a knowledge base and everything in it.”
Publisher declares: read-only no · destructive yes · open-world no
3 inputs, 1 required
- kbIdstringrequired“Knowledge base id, from workbench_kb_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_kb_document_deletedeletesdescribed as changing data
“Removes one ingested document and all its chunks; searches stop returning it at once.”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- kbIdstringrequired“Knowledge base id.”
- documentIdstringrequired“Document id, from workbench_kb_documents_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_kb_listreads or searches
“Lists every knowledge base in the active workspace the caller can see.”
Publisher declares: read-only yes · destructive no · open-world no
1 input
- workspacestring“Workspace slug.”
workbench_kb_searchreads or searches
“Retrieves the most relevant chunks from one knowledge base — the core RAG primitive.”
Publisher declares: read-only yes · destructive no · open-world no
5 inputs, 2 required
- kbIdstringrequired“Knowledge base id (from workbench_kb_list).”
- querystringrequired“What to search for.”
- modestring“semantic (embed + rank by meaning, default) or keyword (free substring match).”
- limitinteger“Max hits to return, 1-50.”
- workspacestring“Workspace slug.”
workbench_kb_createcreates or changes datadescribed as changing data
“Creates an empty knowledge base.”
Publisher declares: read-only no · destructive no · open-world no
6 inputs, 1 required
- visibilitystring“Who can see it: PRIVATE (only the user), WORKSPACE (every member, the default), or SHARED (specific people, granted afterwards).”
- namestringrequired“Knowledge base name (1-200 chars).”
- descriptionstring
- recipestring“Default 'docs'.”
- workspacestring“Workspace slug override.”
- approvalIdstring“Approval id from a prior needs_confirmation envelope.”
workbench_kb_add_documentscreates or changes datadescribed as changing data
“Adds documents to a knowledge base and starts ingestion (chunk + embed; graph KBs also extract entities; tabular KBs load CSV as tables).”
Publisher declares: read-only no · destructive no · open-world no
4 inputs, 2 required
- kbIdstringrequired“Knowledge base id (from workbench_kb_list or workbench_kb_create).”
- sourcesarrayrequired“Documents to ingest (1-50).”
- workspacestring“Workspace slug override.”
- approvalIdstring“Approval id from a prior needs_confirmation envelope.”
workbench_shim_listreads or searches
“Every shim the user can see: id, name, the question it answers, how many answers and examples it has, and the published version's held-out accuracy (null until something is published).”
Publisher declares: read-only yes · destructive no · open-world no
1 input
- workspacestring“Workspace slug.”
workbench_shim_getreads or searches
“One shim in full: the question, where its inputs come from, every answer with its description and examples, what the shim still needs before it can build, and the newest build's report — held-out acc…”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- shimIdstringrequired“Shim id, from workbench_shim_list.”
- workspacestring“Workspace slug.”
workbench_shim_createcreates or changes datadescribed as changing data
“Creates a shim from a name, the question it answers, and its answers.”
Publisher declares: read-only no · destructive no · open-world no
6 inputs, 3 required
- namestringrequired“A short name, e.g.”
- questionstringrequired“The question, e.g.”
- inputsComeFromstring“One line on where the inputs come from, in the user's words.”
- answersarrayrequired
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_shim_add_answercreates or changes datadescribed as changing data
“Adds one answer (a label and, optionally, when it applies).”
Publisher declares: read-only no · destructive no · open-world no
5 inputs, 2 required
- shimIdstringrequired“Shim id, from workbench_shim_list.”
- answerstringrequired
- descriptionstring“When this answer applies.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_shim_add_examplescreates or changes datadescribed as changing data
“Adds example inputs to one answer.”
Publisher declares: read-only no · destructive no · open-world no
5 inputs, 3 required
- shimIdstringrequired“Shim id, from workbench_shim_list.”
- answerstringrequired“The answer these examples belong to, exactly as labelled.”
- examplesarrayrequired
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
comments_liston 6 other serversreads or searches
“Lists the comment threads on one workspace entity (open first, then resolved) with authors and timestamps.”
Publisher declares: read-only yes · destructive no · open-world no
3 inputs, 2 required
- entityKindstringrequired“What the thread hangs on.”
- entityIdstringrequired
- workspacestring“Workspace slug.”
comments_createon 5 other serversdescribed as changing data
“Posts a comment on a workspace entity — a new thread, or a reply when rootId is given.”
Publisher declares: read-only no · destructive no · open-world no
7 inputs, 3 required
- entityKindstringrequired“What the thread hangs on.”
- entityIdstringrequired
- bodystringrequired
- rootIdstring“Reply into this thread; omit to start a new one.”
- mentionedUserIdsarray
- workspacestring“Workspace slug.”
- approvalIdstring
comments_resolveon 5 other serverscreates or changes datadescribed as changing data
“Sets a comment thread's resolved state (rootId = the thread's root comment id).”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- rootIdstringrequired
- resolvedbooleanrequired
- workspacestring“Workspace slug.”
- approvalIdstring
workbench_tasks_getreads or searches
“One Task in full: its sentence, status, the latest saved plan (the clauses — which run flows, which the assistant handles, who decides at each gate) with the flow and people names the plan refers to,…”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- taskIdstringrequired“Task id.”
- workspacestring“Workspace slug.”
workbench_tasks_updatecreates or changes datadescribed as changing data
“Changes a task's name — the gallery label only.”
Publisher declares: read-only no · destructive yes · open-world no
4 inputs, 2 required
- taskIdstringrequired“Task id, from workbench_tasks_list.”
- taskNamestringrequired“New name.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_deletedeletesdescribed as changing data
“Deletes a task.”
Publisher declares: read-only no · destructive yes · open-world no
3 inputs, 1 required
- taskIdstringrequired“Task id, from workbench_tasks_list.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_set_sitescreates or changes datadescribed as changing data
“Sets the full list of sites a Task's machine may reach (example.com, or *.example.com for its subdomains; up to 20), for steps that have to use a site like a person would.”
Publisher declares: read-only no · destructive no · open-world yes
4 inputs, 2 required
- taskIdstringrequired“Task id, from workbench_tasks_list.”
- sitesarrayrequired“Every site the Task may reach.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_listreads or searches
“Lists the workspace's Tasks — sentence-orchestrated processes — with their sentence, status, and recent runs (id, status, who a waiting run is on).”
Publisher declares: read-only yes · destructive no · open-world no
1 input
- workspacestring“Workspace slug.”
workbench_tasks_createcreates or changes datadescribed as changing data
“Creates a DRAFT Task from a name + sentence.”
Publisher declares: read-only no · destructive no · open-world no
4 inputs, 2 required
- taskNamestringrequired“Short task name.”
- sentencestringrequired“The process sentence, in the user's words.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_compile
“Runs the compiler over a sentence and returns the PROPOSED plan (or repair-grade problems — fix by refining the sentence with the user, not by guessing).”
Publisher declares: read-only no · destructive no · open-world no
4 inputs, 1 required
- approvalIdstring“Re-call with the approvalId from a needs_confirmation response after the user approves.”
- sentencestringrequired“The sentence to compile.”
- taskIdstring“The task this sentence belongs to, when compiling an edit — keeps it out of the child-task candidates.”
- workspacestring“Workspace slug.”
workbench_tasks_freezecreates or changes datadescribed as changing data
“Freezes a reviewed plan as the Task's next version and activates it.”
Publisher declares: read-only no · destructive yes · open-world no
5 inputs, 3 required
- taskIdstringrequired“The task id (create a draft first if none).”
- sentencestringrequired“The sentence this plan was compiled from.”
- planrequired“The plan object from workbench_tasks_compile, verbatim.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_startdescribed as changing data
“Kicks off a run of a Task and returns at once; the run carries on by itself and is live on the task's page.”
Publisher declares: read-only no · destructive no · open-world yes
4 inputs, 1 required
- taskIdstringrequired“Task id (from workbench_tasks_list).”
- input“Trigger payload — the document text / JSON the task starts from.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
workbench_tasks_run_viewreads or searches
“The run's current state rendered as its sentence: what ran (with timings), which path each gate took, what's waiting on whom and for how long, what's coming up.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs, 1 required
- runIdstringrequired“Task run id.”
- workspacestring“Workspace slug.”
workbench_tasks_actcreates or changes datadescribed as changing data
“Records THIS user's verdict on a run that is waiting on them — the labels come from the gate (workbench_tasks_run_view shows them via the plan; a wrong label errors listing the real options).”
Publisher declares: read-only no · destructive no · open-world yes
5 inputs, 2 required
- runIdstringrequired“Task run id that is waiting_human.”
- verdictstringrequired“One of the gate's verdict labels, e.g.”
- notestring“The user's reasoning, attached to the verdict.”
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”
search_workspaceon 10 other serversreads or searches
“Finds anything in the workspace in one call — uploaded files (searching the text inside them), folders, Workbench flows, shims, knowledge bases and tasks, Compass pages and changes, Caliper datasets,…”
Publisher declares: read-only yes · destructive no · open-world no
5 inputs, 1 required
- qstringrequired“Words, any order, case-insensitive.”
- kindsarray“Restrict to these kinds (file, folder, flow, page, doc, dataset, …).”
- folderIdstring“Only search inside this folder and the folders under it (ids from workspace_folders_browse or a folder hit).”
- limitinteger
- workspacestring“Workspace slug.”
entity_tags_geton 5 other serversreads or searches
“Returns the tags on a batch of entities of one kind — the labels galleries organize by.”
Publisher declares: read-only yes · destructive no · open-world no
3 inputs, 2 required
- entityKindstringrequired“Which kind the ids belong to.”
- idsarrayrequired
- workspacestring“Workspace slug.”
entity_tags_browseon 5 other serversreads or searches
“Without a tag: every tag in use across the workspace with how many entities carry it, most-used first — the vocabulary the team already organizes by.”
Publisher declares: read-only yes · destructive no · open-world no
2 inputs
- tagstring“A tag to expand into its items.”
- workspacestring“Workspace slug.”
entity_tags_seton 5 other serverscreates or changes datadescribed as changing data
“Replaces the FULL tag set on one entity (an empty list clears it).”
Publisher declares: read-only no · destructive yes · open-world no
5 inputs, 3 required
- entityKindstringrequired
- entityIdstringrequired
- tagsarrayrequired
- workspacestring“Workspace slug.”
- approvalIdstring“Approval id from a prior needs_confirmation response.”

Each quoted sentence is the first sentence of the publisher’s own description. The kind of tool and the “changes data” mark are our AI model’s reading of that description, not something we tested: two differently worded runs had to agree, otherwise no label is shown, and nobody has checked them by hand. A tool without the “changes data” mark is described as only reading, searching or calculating, or the two runs did not agree. No tool was called. Read the publisher’s full description before relying on it. “Publisher declares” lists the hints the server itself sends with the tool (MCP annotations: read-only, destructive, idempotent, open-world); they are the publisher's claims, not checked by us. Inputs are the parameters in the tool's own input schema: name, declared type, whether it is required, and the first sentence of its description. Read 2026-10-23.

How well are its tools described?

Features #4 and #5

How clearly this server's own tool descriptions say what a tool does, and how consistent its tool list is. The first part is an AI model's yes/no reading of each description; the second part is counted. We do not turn it into a grade.

Descriptions

AI reading of each description

Says what the tool does: 100% of 61 tools
Says what it returns: 50% of 58 tools
States limits, defaults or errors: 12% of 57 tools
Says when to use it: 84% of 44 tools
Items covered per tool: 2.4 of 4 on average

Counted from the listing

Parameters with a description: 89% of 253
Tools with a description: 100% of 61
Typical description length: 396 characters

The tool set

Counted from the listing

Tools: 61
Size of the list: about 22,243 tokens for an agent to load (characters ÷ 4)
Naming: 61 of 61 names use snake_case

AI reading: easy to mix up

No pair found.

The AI items cover 61 of 61 tools: for a tool where two differently worded runs disagreed, that item is left out. They were not checked by hand, and no tool was called. “Easy to mix up” means two tools whose names or descriptions overlap without saying how they differ; the AI found none, comparing tools within groups of 40. We do not rate whether the tool set is complete or whether side effects are spelled out, because our AI reading of those was not reliable enough.

How do I add it to my client?

Config from the registry entry

From the remote endpoint (Streamable HTTP)

We connected to this URL on 2026-10-23 (a read-only session: no tool was called).

Add to CursorAn install link in the format Cursor documents, holding the Cursor config below.

Claude Desktop or claude.ai

Settings → Connectors → Add custom connector, then paste:
https://api.zerowidth.ai/mcp/x/workbench

Cursor · .cursor/mcp.json in a project, or ~/.cursor/mcp.json

{
  "mcpServers": {
    "workbench": {
      "url": "https://api.zerowidth.ai/mcp/x/workbench"
    }
  }
}

VS Code · .vscode/mcp.json

{
  "servers": {
    "workbench": {
      "type": "http",
      "url": "https://api.zerowidth.ai/mcp/x/workbench"
    }
  }
}

VS Code (command line) · Bash or zsh terminal (macOS, Linux)

code --add-mcp '{"name":"workbench","type":"http","url":"https://api.zerowidth.ai/mcp/x/workbench"}'

Claude Code

claude mcp add --transport http workbench https://api.zerowidth.ai/mcp/x/workbench

Try it in the MCP Inspector · runs on your machine, opens in your browser · Node 22.19+

npx @modelcontextprotocol/inspector@2.11.0 --server-url https://api.zerowidth.ai/mcp/x/workbench --transport http

Optional settings the entry also lists (left out above): Authorization.
The server runs on the operator's infrastructure; it may ask you to sign in when you first connect.
The MCP Inspector lists the server's tools, resources and prompts; a tool you call there runs for real.

Built from the server's registry entry with each client's documented format (checked 2026-10-14; the Cursor link and the VS Code command 2026-10-19; the MCP Inspector command 2026-10-21). Values in < > are yours to fill in; this page never contains a key. Where the server needs a setting the entry does not declare, it can still fail to start.

Does the registry listing match what we found?

Feature #6

The registry verifies a package once, when it is published, and does not re-check it. A difference below is about the listing, not a judgement of the publisher.

Registry says	We found	Result	Source · date
Remote endpoint api.zerowidth.ai/mcp/x/workbench	Session opened	matches	our test · 2026-10-23
No credential declared as required	Opened a session without any credential	matches	our test · 2026-10-23

Choosing

Record checks

Public records · not a safety rating

1 of 2 record checks met · 7 do not apply

Not ranked by downloads: a remote server publishes no download figure
Not ranked by stars: no public GitHub repository

MetRegistry entry is active and says how to run itActive, with a package or a remote address.
Not metSource repository link opensThe registry entry gives no repository link.
Does not applyRepository is not archivedNo public GitHub repository to read.
Does not applyCode pushed to the repository in the 90 days before our checkNo public GitHub repository to read.
Does not applyRepository publishes a security policy (SECURITY.md)No public GitHub repository to read.
Does not applyPackage names its licence with a standard (SPDX) identifierRemote server: it ships no package.
Does not applyThe version the registry lists can be installed and is not withdrawnRemote server: it ships no package.
Does not applynpm package carries build provenance (attestation)Remote server: it ships no package.
Does not applyNo known critical or high advisory in the package or its dependenciesRemote server: it ships no package.

Read from public records only: the registry entry, npm/PyPI, GitHub and advisory databases, checked 2026-10-22. No tool was called. Only “met” and “not met” are counted, so the total differs between servers; a check that does not apply or that we could not run is listed, never counted against it. It is not a safety rating. How each check works · Leaderboards

Common questions about ai.zerowidth/workbench

Answered from the data above

Is ai.zerowidth/workbench safe to install?

We do not rate servers as safe or unsafe. 1 of its 2 applicable record checks are met (facts from public records, each listed on this page). What the records show: 34 of its tools describe changing data (our AI model's reading of the descriptions). These are facts with their sources on this page, not a safety guarantee, and we did not run its tools.

Is ai.zerowidth/workbench free to use?

Its website (zerowidth.ai/docs/overview/account/plans-and-billing) describes a freemium pricing model. This is an AI-classified reading of the page on 2026-10-23, so check the page itself before relying on it.

Does ai.zerowidth/workbench need an API key or sign-in?

It opened a session without any credential on 2026-10-23. The registry entry declares no required credential; no declaration is not a guarantee.

Has ai.zerowidth/workbench been security tested?

Partly. 4 of the 15 checks that apply to it were exercised. It accepted a session without any credential and lists tools whose names suggest they change data. We did not call them. No security verdict is published before verification.

What can ai.zerowidth/workbench do?

It lists 61 tools. In our AI model's reading of their descriptions (two runs had to agree, nobody checked it by hand), they include tools that read or search, create or change data, delete and run code or commands. We did not call any tool, so this describes the text, not the behaviour.

Can ai.zerowidth/workbench change or delete data?

34 of its tools say in their description that they create, edit, delete, send or run something (our AI model's reading). We did not call any tool, so we cannot say what actually happens.

Found this page useful?

Anonymous · no sign-in

A quick, anonymous reaction — no account, no email. One per kind from your browser. The tally updates when the page is next refreshed, and it is only a rough count, never a safety rating or a measured number of people.

Similar servers

Same topic · AI-read

io.github.callstack/agent-deviceagent-devicealso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one
io.github.A9T9/uivision-mcpUi.Vision MCPalso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one
io.github.Buoy-gg/mcpalso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one
io.github.jumodada/drissionpage-mcpDrissionPage MCPalso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one
io.github.tugcantopaloglu/godot-mcpGodot MCPalso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one
io.github.tomyud1/godot-mcpalso has tools that creates or changes data, deletes, runs code or commands, reads or searchescompare with this one

Active servers whose topic is also “developer-tools” and that have at least one tool of the same kind; most kinds in common first, then most GitHub stars. The topic and the tool kinds are an AI model's reading of the publishers' descriptions, not checked by hand. Similar is not a recommendation.

Compare with another server

Side by side: registry entry, our read-only session, the tools and the 18 classification fields. Same values as each page; no score.

Security testing

What did we test, and what did we leave out?

18 checks · features #1, #2

Open without sign-in

It accepted a session without any credential and lists tools whose names suggest they change data. We did not call them. No security score is given; a check we did not run is never counted as passed.

The 18 checks at a glanceour test · 2026-10-23

Inventory: exercised
Hidden instructions: exercised
Real execution: not run · our policy
Secret leak: not run · nothing to check it against
Internal addresses: not run · our policy
Unconfirmed changes: not run · our policy
Bad input: not run · our policy
Model requests: not run · nothing to check it against
Sensitive questions: not run · nothing to check it against
Folder boundary: not run · our policy
Log leaks: not run · nothing to check it against
Hidden content: exercised
Fake assistant turns: not applicable
Token passthrough: not applicable
Token audience: not applicable
Sign-in redirect: not run · our policy
Consent handling: not testable from outside
Sign-in metadata: exercised

4 exercised11 not run (reason given)3 not applicable

Each check ends in exactly one state: exercised, not run (with the reason), or not applicable (with the reason). 4 of the 15 checks that apply to it were exercised. We do not publish security-check outcomes for named servers.

Exercised · 4

1Inventory
tool-listReads the declared tools, resources and prompts.exercised
2Hidden instructions
description-poisoningLooks for hidden instructions in tool descriptions.exercised
12Hidden content
audience-hidingChecks resources that hide content from the user.exercised
18Sign-in metadata
oauth-url-schemeChecks the published sign-in metadata for unsafe URLs.exercised

Not run, with the reason · 11

3Real execution
real-executionCalls a tool and checks what really happens.This check calls the server's tools. ProtocolProbe calls no server's tools for anything it publishes; an operator can run this check on their own server with our self-test command.not run · our policy
4Secret leak
canary-leakPlants a fake secret and checks whether it leaks.This check needs at least one successful tool call to have something to check, and no tool was called.not run · nothing to check it against
5Internal addresses
ssrfChecks whether a tool can be steered to internal addresses.This check calls the server's tools. ProtocolProbe calls no server's tools for anything it publishes; an operator can run this check on their own server with our self-test command.not run · our policy
6Unconfirmed changes
write-without-confirmationChecks whether tools that change data act without asking.This check calls the server's tools. ProtocolProbe calls no server's tools for anything it publishes; an operator can run this check on their own server with our self-test command.not run · our policy
7Bad input
error-handlingSends bad input and watches how the server answers.This check calls the server's tools. ProtocolProbe calls no server's tools for anything it publishes; an operator can run this check on their own server with our self-test command.not run · our policy
8Model requests
samplingChecks whether the server asks the client's model to act for it.This check needs at least one successful tool call to have something to check, and no tool was called.not run · nothing to check it against
9Sensitive questions
elicitation-sensitive-dataChecks whether the server asks users for sensitive data.This check needs at least one successful tool call to have something to check, and no tool was called.not run · nothing to check it against
10Folder boundary
root-boundaryChecks whether the server reaches outside the folders it was given.This check calls the server's tools. ProtocolProbe calls no server's tools for anything it publishes; an operator can run this check on their own server with our self-test command.not run · our policy
11Log leaks
logging-side-channelChecks whether log messages carry data they should not.This check needs at least one successful tool call to have something to check, and no tool was called.not run · nothing to check it against
16Sign-in redirect
open-redirectChecks the sign-in redirect for abuse.Testing the sign-in redirect means sending probes to the operator's sign-in service, which we do not do.not run · our policy
17Consent handling
confused-deputyChecks consent handling when one service acts for another.Depends on the operator's own consent screen, which differs per server and cannot be tested from outside.not testable from outside

Does not apply to this server · 3

13Fake assistant turns
fake-assistant-turnChecks prompts that pose as the assistant's own words.The server exposes no prompts.not applicable
14Token passthrough
token-passthroughChecks whether the server forwards a user's token to other services.The server opened a session without sign-in, so there is no token flow to check.not applicable
15Token audience
token-audience-validationChecks whether tokens meant for another service are accepted.The server opened a session without sign-in, so there is no token flow to check.not applicable

Source: our own test, 2026-10-23, mcp-runtime-engine@0.1.0. Which checks count and why follows our published scoring model.

What happened during the test?

Feature #8

The steps in order, as the test record holds them.

0 msStarted a read-only session with the server's remote endpoint
848 msConnected after 0 redirects. No credential was sent or needed.
Read the inventory: 61 tools, 134 resources, 0 prompts.
15,530 msFinished: 4 checks exercised.

Reference

How is it classified?

18 fields

One value per field, each with its source. Fields written by AI models are marked. A field with no value says why.

TopicAI-classified

developer-tools

AI-classified · 2026-10-23

Pricing modelAI-classified

freemium

read from zerowidth.ai/docs/overview/account/plans-and-billing · 2026-10-23

“There are four tiers — Free , Starter , Pro , and Enterprise .”

Programming language

does not apply

no existing GitHub repository (server declares no repository link, so there is nothing to fetch)

Hosting type

remote (hosted by the publisher)

registry · 2026-10-23

MCP capability

resources, tools

our test · 2026-10-23

Transport

streamable HTTP

registry · 2026-10-23

Required credentials

none declared

registry · no declaration is not a guarantee

Package registry

does not apply

remote-only server: no package is shipped, so there is no package registry

Freshness

0–30 days

registry updatedAt · 2026-10-23

Lifecycle status

active

registry · 2026-10-23

License class

does not apply

remote-only server: ships no package, so there is nothing to look up in a package registry

Integrity hash

does not apply

remote-only server: ships no package, so there is nothing to look up in a package registry

Signature / provenance

does not apply

remote-only server: ships no package, so there is nothing to look up in a package registry

Dependency advisories

does not apply

ships no package, so there are no dependencies to audit

Popularity

does not apply

remote-only server: ships no package, so there is nothing to look up in a package registry

Publisher signals

not stated

no readable GitHub owner to describe

Maintenance

does not apply

server declares no repository link, so there is nothing to fetch

Environment

the publisher's infrastructure

where it runs, from the hosting type

“Not stated” means the source did not state it; “does not apply” means it does not apply to this kind of server; “our limit” means our own lookup failed. A linked value opens the directory filtered to the same value.

Cite or correct this page

Sources and corrections

Suggested citation

ProtocolProbe. “ai.zerowidth/workbench: MCP server record.” Data checked 2026-10-23. https://protocolprobe.com/mcp/servers/ai.zerowidth/workbench

Figures are valid for the date shown. Our facts on this page are CC BY 4.0: reuse them with a link and the check date. Quoted text (the publisher’s descriptions, website quotes, tool names) is not ours to license.

As JSON: this record in the API · API documentation.

Badge for your README

[![ProtocolProbe](https://protocolprobe.com/api/v1/badge/ai.zerowidth/workbench)](https://protocolprobe.com/mcp/servers/ai.zerowidth/workbench)

It shows this server's record-check count: facts from public records, no tool called. It is not a security rating or a pass. It links here and updates after each data reload.

Something wrong, or is this yours?

If a fact here is wrong, tell us and we will re-check it. If you operate this server, you can run our checks on it yourself with the free self-test command; we call no server’s tools ourselves for anything shown here.

Write to hello@protocolprobe.com · corrections log · methodology