Parallel

Parallel MCP integration

Starts schema-driven research runs, builds and enriches FindAll entity lists, batches runs into task groups, and monitors queries on a cadence.

32actions available

Three actions you can hand over today

Every action runs live through MCP. Nothing to build, nothing to maintain.

  • Create chat completions

    Lyro generates AI-powered responses in real-time to answer customer questions or draft content on the fly. This accelerates response times without human delay.

  • Retrieve task run result

    Lyro fetches the results of a background task so you know when scrapes, analyses, or batch jobs complete. This eliminates polling and keeps work flowing.

  • Extract content from urls

    Lyro pulls relevant text from URLs to load context about a customer's site, competitor, or referenced article instantly. This enriches conversations with live data.

See all 32 actions

How businesses use Parallel + Lyro

Each card is one request a support team gets, and the Parallel actions Lyro runs to close it.

  • Turn a research question into structured output

    Lyro drafts a task specification from a plain description of what you want, creates the run, and returns the finished result against the schema you asked for rather than a page of prose.

    Suggest TaskCreate Task RunRetrieve Task Run Result
  • Build a list of every entity that matches

    Lyro turns a natural-language objective into a FindAll specification, starts the run with its match conditions and exclusions, and fetches the matched and unmatched candidates when it settles.

    Ingest FindAll RunStart FindAll RunGet FindAll Run Result
  • Run a batch of research tasks together

    Lyro creates a task group, launches many runs inside it in parallel, and streams the group's runs back with their inputs and outputs so a hundred-row research job is tracked as one job.

    Create Task GroupAdd Runs to Task GroupFetch Task Group Runs
  • Watch a query and react when something changes

    Lyro sets up a monitor that re-runs a query on a chosen cadence, reads back the errors and material changes it has recorded, and adjusts the cadence or webhook target when the brief moves.

    Create MonitorList Monitor EventsUpdate Monitor

How it works

Get started in 3 steps

Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Lyro reads the Parallel actions it has and picks the ones a request needs.

  1. 01

    Connect Parallel

    Authorize the Parallel account your team already uses — one consent screen, no API keys, no mapping tables. Lyro can only do what you granted that account, and you can disconnect it at any time.

  2. 02

    Tell your agent what you need

    Describe the job the way you would hand it to a teammate. Lyro maps it to the Parallel actions that close it and chains as many as the request needs.

  3. 03

    Watch it work

    The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.

    Get started free
Parallel · Lyro

Everything else about Parallel

Setup, permissions, and the limits of what Lyro can do inside Parallel.

  • Not unless you want it to. Create Task Run returns straight away with the run in a queued state, Retrieve Task Run blocks nothing and just reports status, and Retrieve Task Run Result is the call that waits for completion. Stream Task Run Events gives progress updates in between, so Lyro can report back partway through.

Every action available in Parallel

All 32 actions your agent can call on Parallel, straight from the live MCP connection.

  • Add enrichment to findall run

    Add an enrichment to a FindAll run.

  • Add runs to task group

    Initiate multiple task runs within a TaskGroup.

  • Cancel findall run

    Cancel an active FindAll run by findall_id.

  • Create chat completions

    Get realtime chat completions from Parallel AI.

  • Create monitor

    Create a web monitor that periodically runs the specified query.

  • Create task group

    Create a new task group.

  • Create task run

    Create and initiate a task run.

  • Delete monitor

    Delete a monitor, stopping all future executions.

  • Extend findall run

    Extend a FindAll run by adding additional matches to the current match limit.

  • Extract content from urls

    Extract relevant content from specific web URLs.

  • Fetch task group runs

    Retrieve task runs from a Task Group as a resumable stream.

  • Start findall run

    Start a FindAll run.

  • Get findall run result

    Fetch the final (or latest available) FindAll candidates and result payload for a run.

  • Get findall run schema

    Retrieve the schema configuration of a FindAll run by findall_id.

  • Ingest findall run

    Transform a natural language search objective into a structured FindAll specification.

  • List monitor events

    List events for a monitor from up to the last 300 event groups.

  • List monitors

    List active monitors for the user.

  • Retrieve event group

    Retrieve an event group for a monitor.

  • Retrieve findall run status

    Retrieve status and metadata for a FindAll run by findall_id.

  • Retrieve monitor

    Retrieve a specific monitor by ID.

  • Retrieve task group

    Retrieve details of a specific task group.

  • Retrieve task group run

    Retrieve run status by run_id for a task group.

  • Retrieve task run

    Retrieve run status by run_id.

  • Retrieve task run input

    Retrieve the input data of a specific task run by run_id.

  • Retrieve task run result

    Retrieve the result of a task run by run_id, blocking until the run completes.

  • Parallel search

    Perform parallel semantic search.

  • Simulate event

    Simulate sending an event for a monitor.

  • Stream findall events

    Stream events from a FindAll run.

  • Stream task group events

    Stream events for a Task Group.

  • Stream task run events

    Stream events for a Task Run.

Ready to connect Parallel?

Authorize the account and your agent has all 32 actions from the first conversation.

Support agent working at a laptop next to the Lyro mascot