Skip to content

Connectors · AI

Parallel Web Systems MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Parallel Web Systems and the rest of your systems. It gives your agent 33 Parallel Web Systems actions (18 reads, 11 writes and 4 that wait for a confirmation) and holds the Parallel Web Systems credential itself, so the agent never does. It paces Parallel Web Systems's limits and records every call.

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

claude mcp add --transport http actionlake https://connect.actionlake.com/mcp
the first call
do {  action: "parallel_web.monitors.list",  intent: "check which Parallel Web Systems account is connected"}# List monitors · read · 1 upstream call
  • 33actionsin one signed file
  • 18readlook, never change
  • 11writecreate and update
  • 4wait for a yesdestructive, confirmed first

Set up

Connect Parallel Web Systems in three steps.

  1. 01

    Add actionlake to Claude.

    Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.

    https://connect.actionlake.com/mcp
  2. 02

    Connect Parallel Web Systems.

    API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.

    • api_keyParallel API key: platform.parallel.ai > Settings > API Keys
    account { op: "connections.requirements", service: "parallel_web" }
    # free: exactly what to paste, and what is found for you
  3. 03

    Make the first call. Check before a write.

    Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

    do    { action: "parallel_web.monitors.list" }# list monitors · read · 1 upstream callcheck { action: "parallel_web.task_runs.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Parallel Web Systems at a glance

Sign-in
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
You provide
api_key
Limits we pace to
600 calls a minute, per connection
API covered
Parallel API 0.1.2 (OpenAPI 3.1.0)api.parallel.ai
Built from
Parallel Web Systems's documentation, read 23 October 2026docs.parallel.aidocs.parallel.ai
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
7 of 18 reads, for up to 5 minutes; a write clears what it affects
Shared entities
1, so ask can line Parallel Web Systems up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Parallel Web Systems.

33 actions in 11 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 18readlook, never change
  • 11writecreate and update
  • 4destructivewait for a confirmation
  1. Monitors

    8

    4 read · 3 write · 1 destructive

    List monitors · Get monitor statistics · Create a monitor · List a monitor's events

  2. Findall runs

    7

    3 read · 3 write · 1 destructive

    Create a FindAll run · Add an enrichment to a FindAll run · Extend a FindAll run's match limit · Get a FindAll run's status

  3. Task groups

    4

    2 read · 2 write · 0 destructive

    Add task runs to a task group · Create a task group · Get a task group · Get a run in a task group

  4. Task runs

    4

    3 read · 1 write · 0 destructive

    Create a task run · Get a task run's status · Get a task run's result (waits for completion) · Get a task run's input

  5. Memory

    3

    1 read · 0 write · 2 destructive

    Retrieve memories · Clear memory for a scope · Evict one task, monitor or FindAll source from memory

  6. Findall

    2

    2 read · 0 write · 0 destructive

    Fast entity search (people or companies) · Turn an objective into a FindAll run spec

  7. Chat

    1

    0 read · 1 write · 0 destructive

    Create a chat completion (web-grounded)

  8. Extract

    1

    1 read · 0 write · 0 destructive

    Extract content from URLs

3 more areas · 3 actions
  • Images 1
  • Responses 1
  • Search 1

Top actions

The Parallel Web Systems API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 1 of Parallel Web Systems's 1 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
parallel_web.searchSearch the webSearch.readplatform.search_result.search18 connectors
parallel_web.extractExtract content from URLsExtract.readthis connector
parallel_web.monitors.listList monitorsList Monitors.readthis connector
parallel_web.task_runs.createCreate a task runCreate Task Run.writethis connector
parallel_web.findall.entity_searchFast entity search (people or companies)Fast Entity Search.readthis connector
parallel_web.task_runs.getGet a task run's statusRetrieve Task Run.readthis connector
parallel_web.monitors.statsGet monitor statisticsMonitor Stats.readthis connector
parallel_web.images.searchSearch for imagesImage search.readthis connector
parallel_web.findall.ingestTurn an objective into a FindAll run specIngest FindAll Run.readthis connector
parallel_web.chat.completionsCreate a chat completion (web-grounded)Chat Completions.writethis connector
parallel_web.responses.createCreate a response (web research agent)Create Response.writethis connector
parallel_web.memory.clearClear memory for a scopeClear Memory.destructivethis connector

Try it

Prompts to paste into Claude.

Each one names the Parallel Web Systems actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    List the companies in Parallel Web Systems and flag likely duplicates by name or domain.

    callsparallel_web.findall.entity_search

  2. Writewrite

    Create a task run in Parallel Web Systems. Show me exactly what you will send before you save it.

    callsparallel_web.task_runs.create

Governance for Parallel Web Systems

Destructive actions wait for a yes.

4 Parallel Web Systems actions are destructive. These are all of them:

  • Clear memory for a scopeparallel_web.memory.clear
  • Evict one task, monitor or FindAll source from memoryparallel_web.memory.evict
  • Cancel a monitorparallel_web.monitors.cancel
  • Cancel a FindAll runparallel_web.findall_runs.cancel

A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

doparallel_web.monitors.cancelconfirm_policy: human

Cancel a monitor?

intent: the user asked me to cancel a monitor

run suspended · waiting for a named person · key parallel-web-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "parallel-web-agent",  allowed_actions: ["parallel_web.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "parallel_web.memory.clear", approver: "[email protected]" },    { match: "parallel_web.memory.evict", approver: "[email protected]" },    { match: "parallel_web.monitors.cancel", approver: "[email protected]" }  ]}
Illustrative. Scoped to Parallel Web Systems only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Parallel Web Systems costs on actionlake.

Every Parallel Web Systems action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Parallel Web Systems refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

  • 1upstream call per actionall 33 Parallel Web Systems actions
  • 2actions pageeach page of a long list is one call
  • 7reads can be cacheda cache hit costs nothing
  • 100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Parallel Web Systems action.

33 actions by area: the id your agent calls, what it does, and its effect.

Monitors8
  • parallel_web.monitors.cancelCancel a monitordestructive
  • parallel_web.monitors.createCreate a monitorwrite
  • parallel_web.monitors.getGet a monitorread
  • parallel_web.monitors.listList monitorsread
  • parallel_web.monitors.list_eventsList a monitor's eventsread
  • parallel_web.monitors.statsGet monitor statisticsread
  • parallel_web.monitors.triggerTrigger a monitor run nowwrite
  • parallel_web.monitors.updateUpdate a monitorwrite
Findall runs7
  • parallel_web.findall_runs.cancelCancel a FindAll rundestructive
  • parallel_web.findall_runs.createCreate a FindAll runwrite
  • parallel_web.findall_runs.enrichAdd an enrichment to a FindAll runwrite
  • parallel_web.findall_runs.extendExtend a FindAll run's match limitwrite
  • parallel_web.findall_runs.getGet a FindAll run's statusread
  • parallel_web.findall_runs.get_resultGet a FindAll run's candidatesread
  • parallel_web.findall_runs.get_schemaGet a FindAll run's schemaread
Task groups4
  • parallel_web.task_groups.add_runsAdd task runs to a task groupwrite
  • parallel_web.task_groups.createCreate a task groupwrite
  • parallel_web.task_groups.getGet a task groupread
  • parallel_web.task_groups.get_runGet a run in a task groupread
Task runs4
  • parallel_web.task_runs.createCreate a task runwrite
  • parallel_web.task_runs.getGet a task run's statusread
  • parallel_web.task_runs.get_inputGet a task run's inputread
  • parallel_web.task_runs.get_resultGet a task run's result (waits for completion)read
Memory3
  • parallel_web.memory.clearClear memory for a scopedestructive
  • parallel_web.memory.evictEvict one task, monitor or FindAll source from memorydestructive
  • parallel_web.memory.retrieveRetrieve memoriesread
Findall2
  • parallel_web.findall.entity_searchFast entity search (people or companies)read
  • parallel_web.findall.ingestTurn an objective into a FindAll run specread
Chat1
  • parallel_web.chat.completionsCreate a chat completion (web-grounded)write
Extract1
  • parallel_web.extractExtract content from URLsread
Images1
  • parallel_web.images.searchSearch for imagesread
Responses1
  • parallel_web.responses.createCreate a response (web research agent)write
Search1
  • parallel_web.searchSearch the webread

Questions

The Parallel Web Systems MCP server, answered.

Is there a Parallel Web Systems MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 33 Parallel Web Systems actions across monitors, findall runs, task groups and task runs, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Parallel Web Systems.

What do I need to connect Parallel Web Systems?

Parallel API key: platform.parallel.ai > Settings > API Keys. The free call account { op: "connections.requirements", service: "parallel_web" } lists exactly what to paste.

Can an AI agent delete or change things in Parallel Web Systems?

Only if its key allows it, and only with an explicit confirmation. 4 Parallel Web Systems actions are destructive, including clear memory for a scope, evict one task, monitor or FindAll source from memory and cancel a monitor. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my Parallel Web Systems API limit?

actionlake paces every call to the limits in the Parallel Web Systems service file: 600 calls a minute per connection. Repeated reads are answered from cache (7 Parallel Web Systems actions can be), which never reaches Parallel Web Systems.

What does a Parallel Web Systems action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (2 Parallel Web Systems actions page). Cache hits, searches, dry runs and calls Parallel Web Systems refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

Which Parallel Web Systems API does it cover?

The Parallel API 0.1.2 (OpenAPI 3.1.0) at api.parallel.ai. It is built from Parallel Web Systems's documentation as read on 23 October 2026. 33 actions in 11 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Parallel Web Systems data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Parallel Web Systems credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

Which AI clients can use it?

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

Give your agent Parallel Web Systems today.

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.