Skip to content

Connectors · Accounting & finance

fal.ai MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to fal.ai and the rest of your systems. It gives your agent 86 fal.ai actions (47 reads, 27 writes and 12 that wait for a confirmation) and holds the fal.ai credential itself, so the agent never does. It paces fal.ai'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: "fal_ai.assets.list",  intent: "check which fal.ai account is connected"}# Browse assets · read · 1 upstream call
  • 86actionsin one signed file
  • 47readlook, never change
  • 27writecreate and update
  • 12wait for a yesdestructive, confirmed first

Set up

Connect fal.ai 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 fal.ai.

    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_keyfal API key: fal.ai dashboard > Keys > Add key (shown once, `[key_id]:[secret]`)
    account { op: "connections.requirements", service: "fal_ai" }
    # 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: "fal_ai.assets.list" }# browse assets · read · 1 upstream callcheck { action: "fal_ai.asset_collections.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

fal.ai 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
Fal Model APIs (queue, sync) + Platform API v1api.fal.ai
Built from
fal.ai's documentation, read 23 October 2026api.fal.aifal.aifal.ai
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
45 of 47 reads, for up to 5 minutes; a write clears what it affects
Shared entities
2, so ask can line fal.ai up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in fal.ai.

86 actions in 20 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 47readlook, never change
  • 27writecreate and update
  • 12destructivewait for a confirmation
  1. Asset collections

    12

    4 read · 6 write · 2 destructive

    List asset collections · Get asset collection · Create asset collection · Update asset collection

  2. Models

    8

    7 read · 1 write · 0 destructive

    Billing Events · Estimate cost · Published model insights and weekly statistics (beta) · Model search

  3. Asset characters

    7

    2 read · 4 write · 1 destructive

    List asset characters · Get asset character · Create asset character · Favorite asset character

  4. Asset tags

    7

    1 read · 4 write · 2 destructive

    List asset tags · Create asset tag · Assign tag to asset · Set tags for an asset

  5. Assets

    7

    4 read · 3 write · 0 destructive

    Browse assets · Get asset · List tags for an asset · Get asset lineage

  6. Serverless apps

    6

    5 read · 0 write · 1 destructive

    List Applications · Application Events · Application Revisions · Runner State History

  7. Asset entities

    5

    2 read · 2 write · 1 destructive

    List asset entities · Get asset entity · Create asset entity · Update asset entity

  8. Storage

    5

    2 read · 3 write · 0 destructive

    Get file ACL · Get storage settings · Sign file URL · Set file ACL

12 more areas · 29 actions
  • Keys 4
  • Queue 4
  • Compute instances 3
  • Organization 3
  • Requests 3
  • Serverless files 3
  • Workflows 3
  • Serverless 2
  • Account 1
  • Platform 1
  • Serverless logs 1
  • Serverless requests 1

Top actions

The fal.ai API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 13 of fal.ai's 13 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
fal_ai.assets.listBrowse assetsBrowse assets.readcontent.file.list77 connectors
fal_ai.assets.getGet assetGet asset.readcontent.file.get74 connectors
fal_ai.asset_collections.listList asset collectionsList asset collections.readcontent.folder.list62 connectors
fal_ai.asset_collections.getGet asset collectionGet asset collection.readcontent.folder.get47 connectors
fal_ai.workflows.listList user workflowsList user workflows.readautomation.workflow.list51 connectors
fal_ai.asset_collections.createCreate asset collectionCreate asset collection.writecontent.folder.create62 connectors
fal_ai.keys.listList API KeysList API Keys.readidentity.api_key.list96 connectors
fal_ai.workflows.getGet workflow detailsGet workflow details.readautomation.workflow.get51 connectors
fal_ai.asset_collections.updateUpdate asset collectionUpdate asset collection.writecontent.folder.update48 connectors
fal_ai.asset_collections.deleteDelete asset collectionDelete asset collection.destructivecontent.folder.delete46 connectors
fal_ai.keys.createCreate API KeyCreate API Key.writeidentity.api_key.create100 connectors
fal_ai.organization.teamsOrganization TeamsOrganization Teams.readthis connector

Try it

Prompts to paste into Claude.

Each one names the fal.ai actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    Which fal.ai files changed this week, and who changed them?

    callsfal_ai.assets.list

  2. Writewrite

    Create asset collection in fal.ai. Show me exactly what you will send before you save it.

    callsfal_ai.asset_collections.create

Governance for fal.ai

Destructive actions wait for a yes.

12 fal.ai actions are destructive. These are all of them:

  • Delete asset collectionfal_ai.asset_collections.delete
  • Delete API Keyfal_ai.keys.delete
  • Revoke API Keyfal_ai.keys.revoke
  • Delete asset tagfal_ai.asset_tags.delete
  • Unassign tag from assetfal_ai.asset_tags.unassign
  • Delete asset entityfal_ai.asset_entities.delete
  • Delete asset characterfal_ai.asset_characters.delete
  • Delete Compute Instancefal_ai.compute_instances.delete
  • Delete request payloadsfal_ai.requests.delete_payloads
  • Flush Application Queuefal_ai.serverless_apps.flush_queue
  • Remove asset from collectionfal_ai.asset_collections.remove_asset
  • Cancel a requestfal_ai.queue.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.

dofal_ai.asset_collections.deleteconfirm_policy: human

Delete asset collection?

intent: the user asked me to delete asset collection

run suspended · waiting for a named person · key fal-ai-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "fal-ai-agent",  allowed_actions: ["fal_ai.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "fal_ai.asset_collections.delete", approver: "[email protected]" },    { match: "fal_ai.keys.delete", approver: "[email protected]" },    { match: "fal_ai.keys.revoke", approver: "[email protected]" }  ]}
Illustrative. Scoped to fal.ai only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What fal.ai costs on actionlake.

Every fal.ai action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call fal.ai 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 86 fal.ai actions
  • 86actions pageeach page of a long list is one call
  • 45reads 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 fal.ai action.

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

Asset collections12
  • fal_ai.asset_collections.add_assetAdd asset to collectionwrite
  • fal_ai.asset_collections.createCreate asset collectionwrite
  • fal_ai.asset_collections.deleteDelete asset collectiondestructive
  • fal_ai.asset_collections.favoriteFavorite asset collectionwrite
  • fal_ai.asset_collections.getGet asset collectionread
  • fal_ai.asset_collections.hierarchyGet asset collection hierarchyread
  • fal_ai.asset_collections.listList asset collectionsread
  • fal_ai.asset_collections.list_assetsBrowse assets in a collectionread
  • fal_ai.asset_collections.moveMove asset collectionwrite
  • fal_ai.asset_collections.remove_assetRemove asset from collectiondestructive
  • fal_ai.asset_collections.unfavoriteUnfavorite asset collectionwrite
  • fal_ai.asset_collections.updateUpdate asset collectionwrite
Models8
  • fal_ai.models.analyticsAnalyticsread
  • fal_ai.models.billing_eventsBilling Eventsread
  • fal_ai.models.estimate_costEstimate costread
  • fal_ai.models.insightsPublished model insights and weekly statistics (beta)read
  • fal_ai.models.listModel searchread
  • fal_ai.models.pricingPricingread
  • fal_ai.models.runRun a modelwrite
  • fal_ai.models.usageUsageread
Asset characters7
  • fal_ai.asset_characters.createCreate asset characterwrite
  • fal_ai.asset_characters.deleteDelete asset characterdestructive
  • fal_ai.asset_characters.favoriteFavorite asset characterwrite
  • fal_ai.asset_characters.getGet asset characterread
  • fal_ai.asset_characters.listList asset charactersread
  • fal_ai.asset_characters.unfavoriteUnfavorite asset characterwrite
  • fal_ai.asset_characters.updateUpdate asset characterwrite
Asset tags7
  • fal_ai.asset_tags.assignAssign tag to assetwrite
  • fal_ai.asset_tags.createCreate asset tagwrite
  • fal_ai.asset_tags.deleteDelete asset tagdestructive
  • fal_ai.asset_tags.listList asset tagsread
  • fal_ai.asset_tags.set_for_assetSet tags for an assetwrite
  • fal_ai.asset_tags.unassignUnassign tag from assetdestructive
  • fal_ai.asset_tags.updateUpdate asset tagwrite
Assets7
  • fal_ai.assets.favoriteFavorite assetwrite
  • fal_ai.assets.getGet assetread
  • fal_ai.assets.lineageGet asset lineageread
  • fal_ai.assets.listBrowse assetsread
  • fal_ai.assets.list_tagsList tags for an assetread
  • fal_ai.assets.unfavoriteUnfavorite assetwrite
  • fal_ai.assets.upload_from_urlUpload assetwrite
Serverless apps6
  • fal_ai.serverless_apps.flush_queueFlush Application Queuedestructive
  • fal_ai.serverless_apps.listList Applicationsread
  • fal_ai.serverless_apps.list_eventsApplication Eventsread
  • fal_ai.serverless_apps.list_revisionsApplication Revisionsread
  • fal_ai.serverless_apps.queue_sizeQueue Sizeread
  • fal_ai.serverless_apps.runner_historyRunner State Historyread
Asset entities5
  • fal_ai.asset_entities.createCreate asset entitywrite
  • fal_ai.asset_entities.deleteDelete asset entitydestructive
  • fal_ai.asset_entities.getGet asset entityread
  • fal_ai.asset_entities.listList asset entitiesread
  • fal_ai.asset_entities.updateUpdate asset entitywrite
Storage5
  • fal_ai.storage.get_file_aclGet file ACLread
  • fal_ai.storage.get_settingsGet storage settingsread
  • fal_ai.storage.set_file_aclSet file ACLwrite
  • fal_ai.storage.sign_file_urlSign file URLwrite
  • fal_ai.storage.update_settingsUpdate storage settingswrite
Keys4
  • fal_ai.keys.createCreate API Keywrite
  • fal_ai.keys.deleteDelete API Keydestructive
  • fal_ai.keys.listList API Keysread
  • fal_ai.keys.revokeRevoke API Keydestructive
Queue4
  • fal_ai.queue.cancelCancel a requestdestructive
  • fal_ai.queue.resultGet request resultread
  • fal_ai.queue.statusGet request statusread
  • fal_ai.queue.submitSubmit a requestwrite
Compute instances3
  • fal_ai.compute_instances.deleteDelete Compute Instancedestructive
  • fal_ai.compute_instances.getGet Compute Instanceread
  • fal_ai.compute_instances.listList Compute Instancesread
Organization3
  • fal_ai.organization.billing_eventsOrganization Billing Eventsread
  • fal_ai.organization.teamsOrganization Teamsread
  • fal_ai.organization.usageOrganization Usageread
Requests3
  • fal_ai.requests.delete_payloadsDelete request payloadsdestructive
  • fal_ai.requests.list_by_endpointList requests by endpoint(s)read
  • fal_ai.requests.searchSearch Requestsread
Serverless files3
  • fal_ai.serverless_files.list_directoryList files (directory)read
  • fal_ai.serverless_files.list_rootList files (root)read
  • fal_ai.serverless_files.upload_from_urlUpload file from URLwrite
Workflows3
  • fal_ai.workflows.createCreate a workflowwrite
  • fal_ai.workflows.getGet workflow detailsread
  • fal_ai.workflows.listList user workflowsread
Serverless2
  • fal_ai.serverless.analyticsAnalyticsread
  • fal_ai.serverless.usageUsageread
Account1
  • fal_ai.account.billingAccount Billingread
Platform1
  • fal_ai.platform.metaGet platform metadataread
Serverless logs1
  • fal_ai.serverless_logs.historyLogs history (paginated)read
Serverless requests1
  • fal_ai.serverless_requests.list_by_endpointList requests by endpoint(s)read

Questions

The fal.ai MCP server, answered.

Is there a fal.ai MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 86 fal.ai actions across asset collections, models, asset characters and asset tags, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by fal.ai.

What do I need to connect fal.ai?

Fal API key: fal.ai dashboard > Keys > Add key (shown once, `[key_id]:[secret]`). The free call account { op: "connections.requirements", service: "fal_ai" } lists exactly what to paste.

Can an AI agent delete or change things in fal.ai?

Only if its key allows it, and only with an explicit confirmation. 12 fal.ai actions are destructive, including delete asset collection, delete API Key and revoke API Key. 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 fal.ai API limit?

actionlake paces every call to the limits in the fal.ai service file: 600 calls a minute per connection. Repeated reads are answered from cache (45 fal.ai actions can be), which never reaches fal.ai.

What does a fal.ai action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (86 fal.ai actions page). Cache hits, searches, dry runs and calls fal.ai 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 fal.ai API does it cover?

The fal Model APIs (queue, sync) + Platform API v1 at api.fal.ai. It is built from fal.ai's documentation as read on 23 October 2026. 86 actions in 20 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my fal.ai 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 fal.ai 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 fal.ai today.

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