Small Printthe tool description is the small print. we read it.
mcpnpmmaintained by dan-pplx and 2 others

@perplexity-ai/mcp-server

Its listing says: Real-time web search, reasoning, and research through Perplexity's API

What this server tells an agent last changed in version 1.1.0 on 2026-07-27.

In short

We won't send spam: we email you only when this server changes what it tells your agent, and you can stop at any time.

You need no account and no password. We send one link to confirm first.

@perplexity-ai/mcp-server, an MCP server on npm: 12 versions on record, 4 tools read, 9 changes to the small print across 2 releases, worst change graded medium. Last checked by Small Print on 17 November 2026.

The two tags below are examples showing how a grade is written. They are not grades of anything on this page.

advisoryhigh

An advisory is a security report on record. Its own page names its source.

changehigh

A change is a difference in the small print between two versions. It is graded by a written rule. When a model review lowered a high grade or kept it, the entry page says so beside the change.

Advisories

0 recorded

No advisory on record names this entry. Small Print checks the GitHub Advisory Database, OSV, NVD, the CSA MCP Security vulnerability database and the Vulnerable MCP Project.

Changes to the small print

2 of 10 releases changed it · 9 changes
  1. 0.8.12026-02-16not compared
  2. 0.7.02026-02-16identical
  3. 0.8.22026-02-16identical
  4. 0.8.32026-02-26identical
  5. 0.8.42026-03-20identical
  6. 0.8.52026-03-23identical
  7. 0.9.02026-03-23identical
  8. 1.0.02026-07-277 changes, medium
  9. 1.1.02026-07-272 changes, medium
  10. 1.2.02026-07-27identical
  11. 1.2.12026-08-27identical
  12. 1.3.02026-09-25not compared

no change on recordidenticalnot comparedchanged, graded infochanged, graded lowchanged, graded mediumchanged, graded highchanged, graded criticalnot read yet

1.2.0→1.2.12026-08-27identical small print
1.1.0→1.2.02026-07-27identical small print
1.0.0→1.1.02026-07-27medium1 description, 1 schema
mediumtool.descriptionperplexity_searchdrift.tool.description.changed

The description changed. The added text contains none of the words the instruction, secret or destination rules look for.

- Search the web and return a ranked list of results with titles, URLs, snippets, and dates. Best for: finding specific URLs, checking recent news, verifying facts, discovering sources. Returns formatted results (title, URL, snippet, date) with no AI synthesis. For AI-generated answers with citations, use perplexity_ask instead.
+ Search the web and return a ranked list of results with titles, URLs, snippets, and dates. Best for: finding specific URLs, checking recent news, verifying facts, discovering sources. Returns formatted results (title, URL, snippet, date) with no AI synthesis. Supports recency filters and domain restrictions. For AI-generated answers with citations, use perplexity_ask instead.
lowtool.schemaperplexity_searchdrift.tool.schema.optional-added

The input form the reader found adds a field it does not require, and nothing else in it changed.

  {
    "$schema": "http://json-schema.org/draft-07/schema#",
    "properties": {
      "country": {
        "description": "ISO 3166-1 alpha-2 country code for regional results (e.g., 'US', 'GB')",
        "type": "string"
      },
      "max_results": {
        "description": "Maximum number of results to return (1-20, default: 10)",
        "maximum": 20,
        "minimum": 1,
        "type": "number"
      },
      "max_tokens_per_page": {
        "description": "Maximum tokens to extract per webpage (default: 1024)",
        "maximum": 2048,
        "minimum": 256,
        "type": "number"
      },
      "query": {
        "description": "Search query string",
        "type": "string"
+     },
+     "search_domain_filter": {
+       "description": "Restrict search results to specific domains (e.g., ['wikipedia.org', 'arxiv.org']). Use '-' prefix for exclusion (e.g., ['-reddit.com']).",
+       "items": {
+         "type": "string"
+       },
+       "type": "array"
+     },
+     "search_recency_filter": {
+       "description": "Filter search results by recency. Use 'hour' for very recent news, 'day' for today's updates, 'week' for this week, etc.",
+       "enum": [
+         "hour",
+         "day",
+         "week",
+         "month",
+         "year"
+       ],
+       "type": "string"
      }
    },
    "required": [
      "query"
    ],
    "type": "object"
  }

Graded again on 2026-10-14 under version 10 of the grading rules. Under the earlier rules it was medium (drift.tool.schema).

This page shows the latest 3 releases. Show all 10 releases on one page, or read every change in the read API.

The small print

4 tools in 1.3.0
perplexity_ask
messages · search_context_size? · search_domain_filter? · search_recency_filter?

Answer a question using web-grounded AI (Perplexity Agent API, fast preset). Best for: quick factual questions, summaries, explanations, and general Q&A. Returns a text response with numbered citations. Fastest and cheapest option. Supports filtering by recency (hour/day/week/month/year), domain restrictions, and search context size. For in-depth multi-source research, use perplexity_research instead. For step-by-step reasoning and analysis, use perplexity_reason instead.

perplexity_research
messages

Conduct deep, multi-source research on a topic (Perplexity Agent API, high preset). Best for: literature reviews, comprehensive overviews, investigative queries needing many sources. Returns a detailed response with numbered citations. Significantly slower than other tools (can take minutes). For quick factual questions, use perplexity_ask instead. For logical analysis and reasoning, use perplexity_reason instead.

perplexity_reason
messages · search_context_size? · search_domain_filter? · search_recency_filter?

Analyze a question using step-by-step reasoning with web grounding (Perplexity Agent API, medium preset). Best for: math, logic, comparisons, complex arguments, and tasks requiring chain-of-thought. Returns a reasoned response with numbered citations. Supports filtering by recency (hour/day/week/month/year), domain restrictions, and search context size. For quick factual questions, use perplexity_ask instead. For comprehensive multi-source research, use perplexity_research instead.

Show 1 more tool
perplexity_search
query · country? · max_results? · max_tokens_per_page? · search_domain_filter? · search_recency_filter? · search_type?

Search the web and return a ranked list of results with titles, URLs, snippets, and dates. Best for: finding specific URLs, checking recent news, verifying facts, discovering sources. Returns formatted results (title, URL, snippet, date) with no AI synthesis. Supports recency filters, domain restrictions, and a lower-latency fast search mode. For AI-generated answers with citations, use perplexity_ask instead.

Related

other entries on the record

Versions

12 tracked
versionpublishedfiles fingerprintfilestool text fingerprint
1.3.02026-09-254de397bce844eaab166c5983fb97b47b2a chain 2026-10-04
1.2.12026-08-274d97c1c6bc353b4d160c34d2dd1d7d3ad2 chain 2026-10-04
1.2.02026-07-27not fingerprinted yet0c34d2dd1d7d3ad2 chain 2026-10-04
1.1.02026-07-27not fingerprinted yet0c34d2dd1d7d3ad2 chain 2026-10-04
1.0.02026-07-27not fingerprinted yet3b4db763dfb88c82 chain 2026-10-04
Show 7 older versions
0.9.02026-03-23not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.8.52026-03-23not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.8.42026-03-20not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.8.32026-02-26not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.8.22026-02-16not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.7.02026-02-16not fingerprinted yet0cf793735bd23bde chain 2026-10-04
0.8.12026-02-16not fingerprinted yet0cf793735bd23bde chain 2026-10-06