Skip to main content
Unblocked MCP tools are actively evolving. Tools and their descriptions may change over time.
research_task investigates a broad topic across your connected sources and summarizes what it finds. Your agent uses it when a question spans multiple systems, teams, or sources and no single lookup will give the full picture.

How research differs from the other tools

The other Unblocked MCP tools each answer a specific question: unblocked_context_engine gives context about one topic, data_retrieval returns specific records, and link_resolver fetches the content behind a URL. Each returns one piece of the puzzle. research_task assembles the full picture. It pulls from all of these sources at once and connects them:
  • Why a piece of code was built this way, based on the PRs and discussions that led to it.
  • What patterns exist across multiple files and services.
  • What your team has said in messaging threads, issue trackers, and docs that code search alone would miss.
Your agent uses research when answering a question correctly requires understanding context from more than one place. Getting this upfront means fewer wrong turns and less rework.

When your agent uses it

  • Investigating unfamiliar subsystems or feature areas, such as how the notification pipeline works end-to-end.
  • Questions that span multiple systems, teams, or sources.
  • Scoping an integration, migration, or refactor, such as evaluating the impact of upgrading from React 17 to 18.
  • When it has already tried several lookups without converging on an answer.
  • At the start of a planning task, in parallel with code exploration.

When your agent uses something else