MalleableMalleable

MCP server

Malleable exposes an MCP (Model Context Protocol) server, so Claude and other MCP clients can list, build, edit, and run your workflows from outside the app.

Looking for the other direction, the apps your workflows can connect to? See the integrations directory.

Connecting

  • Endpoint: https://www.usemalleable.com/mcp (streamable HTTP)
  • Authentication: sign in with OAuth 2.0. Point your client at the endpoint and it walks you through a browser sign-in; no key to copy or store. Clients that support MCP's OAuth flow handle this for you.
  • Prefer a token? You can also pass a Malleable API key as a bearer token. Generate one from your account profile (the avatar menu in the top-right of the app).

Either way, tools act as you, so they see and edit the workflows your account can.

Connected, an MCP client can list your workflows, read a workflow's full configuration, build and edit workflows, and start and monitor runs: everything below, acting as your account.

Tools

list-use-casesread-only

List the use cases you can access — your own plus ones shared with your organization — with each one's published and draft workflow ids.

describe-workflowread-only

Read a workflow's structure: its collected data, channels, stages, components, entry points, and sub-workflows; or one stage in full, with each component's configuration and its triggers.

create-use-casereaches outside Malleable

Create a use case from a plain-English description; the architect builds its first workflow asynchronously.

instruct-architectmakes changesreaches outside Malleable

Ask the architect agent to inspect or edit a workflow. It works asynchronously; follow it with get-architect-status.

get-architect-statusread-only

See what the architect is doing on a use case: its status, what it has done since you last checked, and anything waiting on your answer. Can wait for it to finish.

respond-to-architectmakes changesreaches outside Malleable

Answer the architect's pending question or approval. Publishing is the one approval only you can give, in Malleable.

run-executionmakes changesreaches outside Malleable

Start a workflow as a real run or as a test with mocked side effects or a simulated user. Returns at once; follow it with get-execution-status.

get-execution-statusread-only

Check on a run: its status, whether it waits on a person or a hand-off, its collected data, and a link to open it. Can wait for it to settle.

list-executionsread-only

List a use case's recent runs, filtered by status, collected data, or workflow version.

read-execution-dataread-only

Read the data a run collected, in full or one key.

read-execution-diagnosticsread-only

See why a run failed or was rejected: its trigger, failure events, script logs, and the sub-workflow runs it started.

list-execution-snapshotsread-only

List the mid-run snapshots a test can resume from.

list-workflow-versionsread-only

List a use case's workflow versions, newest first.

tally-collected-dataread-only

Count the values one collected-data key has taken across a use case's runs.

message-executionmakes changesreaches outside Malleable

Send a message to a running execution to steer it.

read-fileread-only

Read a malleable:// handle — file uploads, stored tool outputs, webhook payloads, run transcripts — raw, by query, grep, SQL, or range.

search-integrationsread-only

Find integrations workflows can use, whether you have connected each, and a link to connect the ones you haven't.