Skip to content

Proposal: add a Perplexity Search MCP tool as an alternative search source #136

Description

@jliounis

Hi Bright Data team — thanks for maintaining this MCP server! I'd like to propose adding a perplexity_search MCP tool that wraps Perplexity's Search API (POST https://api.perplexity.ai/search) as an alternative/complementary search source alongside search_engine.

Why

The current search_engine tool routes through Bright Data's SERP infrastructure, which is great for grounded SERP scraping. Perplexity's Search API is a different shape — it's a curated, ranked web index optimized for LLM grounding, with first-class filters for domain allow/deny, recency, and date ranges. Exposing it as an additional MCP tool would let agents pick the right tool for the job (e.g., raw SERP vs. LLM-curated index) without leaving the Bright Data MCP.

The existing web_data_perplexity_ai_insights tool scrapes the Perplexity web app via Bright Data — that's a different surface from the Perplexity Search API, which returns structured {title, url, snippet, date} results directly.

Proposed shape

  • Tool name: perplexity_search
  • Auth: reads PERPLEXITY_API_KEY (also accepts PPLX_API_KEY) — independent from API_TOKEN. The tool no-ops / is hidden when neither env var is set, so existing users see no change.
  • Inputs: query (required), optional max_results, search_domain_filter (allow OR deny via -domain.com, never mixed), search_recency_filter (hour|day|week|month|year), search_after_date_filter / search_before_date_filter (m/d/yyyy).
  • Output: array of {title, url, snippet, date}.
  • Group: could live in a new web_search group, or under RESEARCH / ADVANCED_SCRAPING. Open to whatever fits your taxonomy.
  • Tests: mocked HTTP via axios (matches the existing pattern in server.js).
  • Docs: README section + tool reference entry.

Implementation notes

Question for maintainers

Would you accept a PR adding this tool? I'm happy to put one up on a branch (feat/perplexity-search) following your existing conventions in server.js / tool_groups.js / test/ and matching the style of the search_engine tool. If the answer is "this doesn't fit our scope — Bright Data MCP is for Bright Data services only," totally understood and I'll close this out.

Thanks!
— James (james.liounis@perplexity.ai / @james-pplx)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions