Octen Web Search for Source Pages

Give Octen a question or keywords to find ranked webpage results with titles, source links and relevant passages. Choose general or news-focused search, and narrow sources or dates when needed. Your Agent can read the returned material and prepare a cited brief as a separate step.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Use Toolmu's octen-web-search-octen, catalog ID fbf3f3f5-d54f-4fdd-9d65-674a806f44f3, to find sources for my question. Reuse the existing Toolmu connection; if setup is needed, follow https://toolmu.com/SKILL.md. Read get_tool for this exact Tool, check current availability and price, and honor my existing authorization and budget. Ask for the query if it is missing. Apply my chosen result count, topic, country priority, source or text conditions, and date window. Keep publication-based and crawl-based filtering distinct; explicit date bounds take precedence over a relative window. Request highlights or page content at a useful length for the task. Read the actual results and retain their titles, URLs and any supplied author or date context. Keep source wording separate from your analysis, and explain when a passage is too limited to support a conclusion. Prepare my requested brief or comparison as separate Agent work with references to material actually read. Opening further source pages, translating text or fetching image files is additional work; do not present a link alone as a completed reading.

Task:
Find [a plain-language guide to website accessibility] from [w3.org]. Show up to [5 useful pages], keep their links and relevant passages, then summarize what each source says.

Paste into your connected AI's conversation.

Connection guide

Up to 12.0012 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

Octen

Octen Web Search for Source Pages

You receive

The result is a list of source pages relevant to your query, with available titles, URLs and focused text highlights. You can also request available page content or image references; author and date information adds context when supplied. Your Agent can compare the material, remove duplicate references and organize a source-linked answer based on what it actually reads.

Start with your task

Find [a plain-language guide to website accessibility] from [w3.org]. Show up to [5 useful pages], keep their links and relevant passages, then summarize what each source says.

What to know

A search list is not a finished answer or a fact check. Ranking does not prove a source is accurate or that the list covers every relevant page. Highlights and token-limited content can omit context; publication dates and crawl dates describe different events.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Search request`body`	`object`	Required	Required object containing the query and any chosen source, date or content settings. Required fields `query`
Requested result count`body.count`	`integer`	Optional	Request 1–100 results; default 5. This sets the requested list size, not a guarantee that enough matching pages exist. Default `5` Minimum `1` Maximum `100`
Search question or keywords`body.query`	`string`	Required	Required search text, 1–500 characters. State the subject and what you want to find so the returned passages have a useful focus. Minimum length `1` Maximum length `500`
Search topic`body.topic`	`string`	Optional	Choose general for general webpages or news for news-focused results. Default general. News focus does not itself verify a report. Default `general` Options `general, news`
Highlight format`body.format`	`string`	Optional	Choose text or markdown for highlight formatting; default text. This is text presentation rather than an interactive page preview. Default `text` Options `markdown, text`
Country priority`body.country`	`string`	Optional	Prioritize results from an ISO 3166 country. Octen describes auto as inferring the priority from the query when omitted. It does not choose the result language.
Time-window end`body.end_time`	`string`	Optional	Optional end of the filter window, expressed as ISO 8601 text. Include a time offset to make your intended boundary clear. Explicit dates take precedence over time_range.
Highlight settings`body.highlight`	`object`	Optional	Optional object controlling whether query-relevant passages are requested and their token budget. Highlights are portions of a source, not a complete answer.
Return relevant passages`body.highlight.enable`	`boolean`	Optional	Set true to request query-relevant highlights; default true. Turn it off if your task does not need those passages. Default `true`
Highlight token limit`body.highlight.max_tokens`	`integer`	Optional	Limit each highlight to 100–20,000 tokens; default 512. A token is a text-processing unit, not a character or word count. Default `512` Minimum `100` Maximum `20000`
Explicit-content filtering`body.safesearch`	`string`	Optional	Choose strict or off; default strict. strict filters explicit/adult content, while off disables this filter. It is not a factual-reliability setting. Default `strict` Options `off, strict`
Time-window start`body.start_time`	`string`	Optional	Optional beginning of the filter window, expressed as ISO 8601 text. Pair it with the intended time_basis; explicit dates take precedence over time_range.
Date used for filtering`body.time_basis`	`string`	Optional	Choose auto, published or crawled; default auto. published uses publication time, while crawled uses the source crawl time. They describe different events. Default `auto` Options `auto, published, crawled`
Relative time window`body.time_range`	`string`	Optional	Choose day, week, month or year, or the aliases d, w, m or y, counting back from now using time_basis. Explicit start_time/end_time takes precedence. Options `day, week, month, year, d, w, m, y`
Text to exclude`body.exclude_text`	`array`	Optional	Optional list of up to 5 strings to exclude from result-page text, with at most 30 characters in each. Keep the conditions specific enough for your task. Maximum items `5`
One excluded text condition`body.exclude_text[]`	`string`	—	One string, at most 30 characters, specifying text that must not appear on the result page. Maximum length `30`
Page-content settings`body.full_content`	`object`	Optional	Optional object controlling whether page content is requested alongside the results and the token limit per result.
Request page content`body.full_content.enable`	`boolean`	Optional	Set true to request available page content with each result; default false. Access conditions and token limits still affect what is available to read. Default `false`
Page-content token limit`body.full_content.max_tokens`	`integer`	Optional	Set the page-content limit to 100–100,000 tokens per result; default 2048. This is a length budget, not a promise to retrieve every part of the page. Default `2048` Minimum `100` Maximum `100000`
Text to require`body.include_text`	`array`	Optional	Optional list of up to 5 strings that must appear in result-page text, with at most 30 characters in each. Maximum items `5`
One required text condition`body.include_text[]`	`string`	—	One string, at most 30 characters, specifying text that must appear on the result page. Maximum length `30`
Include image references`body.include_images`	`boolean`	Optional	Set true to request available image references with results; default false. A reference is not an image file that has been fetched or saved. Default `false`
Domains to exclude`body.exclude_domains`	`array`	Optional	Optional list of up to 1,200 domains to exclude, each at most 60 characters. Octen’s Search reference gives exclusion precedence when a domain is also included. Maximum items `1200`
One excluded domain`body.exclude_domains[]`	`string`	—	One domain string, at most 60 characters, to exclude from the source list. Maximum length `60`
Domains to include`body.include_domains`	`array`	Optional	Optional list of up to 1,200 domains to include as sources, each at most 60 characters. Use domain names rather than whole article URLs. Maximum items `1200`
One included domain`body.include_domains[]`	`string`	—	One domain string, at most 60 characters, to include in the selected sources. Maximum length `60`

View raw JSON

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "required": [
        "query"
      ],
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      },
      "properties": {
        "count": {
          "type": "integer",
          "default": 5,
          "maximum": 100,
          "minimum": 1,
          "description": "Number of results to return (1-100)."
        },
        "query": {
          "type": "string",
          "maxLength": 500,
          "minLength": 1,
          "description": "The search query."
        },
        "topic": {
          "enum": [
            "general",
            "news"
          ],
          "type": "string",
          "default": "general",
          "description": "Use 'general' for general web search, or 'news' for news-focused results."
        },
        "format": {
          "enum": [
            "markdown",
            "text"
          ],
          "type": "string",
          "default": "text",
          "description": "Formatting of highlight outputs."
        },
        "country": {
          "type": "string",
          "description": "ISO 3166 country to prioritize results from; default 'auto' infers it from the query."
        },
        "end_time": {
          "type": "string",
          "description": "End time for filtering results. ISO 8601 format."
        },
        "highlight": {
          "type": "object",
          "properties": {
            "enable": {
              "type": "boolean",
              "default": true,
              "description": "If true, returns query-relevant highlight in each result."
            },
            "max_tokens": {
              "type": "integer",
              "default": 512,
              "maximum": 20000,
              "minimum": 100,
              "description": "Max tokens returned per highlight."
            }
          },
          "additionalProperties": false
        },
        "safesearch": {
          "enum": [
            "off",
            "strict"
          ],
          "type": "string",
          "default": "strict",
          "description": "Explicit/adult content filtering."
        },
        "start_time": {
          "type": "string",
          "description": "Start time for filtering results. ISO 8601 format."
        },
        "time_basis": {
          "enum": [
            "auto",
            "published",
            "crawled"
          ],
          "type": "string",
          "default": "auto",
          "description": "Which time field is used for time filtering."
        },
        "time_range": {
          "enum": [
            "day",
            "week",
            "month",
            "year",
            "d",
            "w",
            "m",
            "y"
          ],
          "type": "string",
          "description": "Relative time window counting back from now. Mutually exclusive with start_time/end_time (which take precedence)."
        },
        "exclude_text": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 30
          },
          "maxItems": 5,
          "description": "Strings that must not appear in the result page text."
        },
        "full_content": {
          "type": "object",
          "properties": {
            "enable": {
              "type": "boolean",
              "default": false,
              "description": "If true, returns full_content for each result."
            },
            "max_tokens": {
              "type": "integer",
              "default": 2048,
              "maximum": 100000,
              "minimum": 100,
              "description": "Maximum tokens of full content included per result."
            }
          },
          "additionalProperties": false
        },
        "include_text": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 30
          },
          "maxItems": 5,
          "description": "Strings that must appear in the result page text."
        },
        "include_images": {
          "type": "boolean",
          "default": false,
          "description": "Whether to include images in each result."
        },
        "exclude_domains": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 60
          },
          "maxItems": 1200,
          "description": "Domains to specifically exclude from the search results."
        },
        "include_domains": {
          "type": "array",
          "items": {
            "type": "string",
            "maxLength": 60
          },
          "maxItems": 1200,
          "description": "Domains to specifically include in the search results."
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Additional notes

Read the results as a source list. A high position signals Octen’s search ranking, not independent confirmation of the page’s claims. Keep the title and URL with the passage you use. Different pages may repeat the same claim, so your Agent can compare sources and remove duplicate references separately.

Highlights focus on the query and are enabled by default. They can help you choose what to read, but do not represent every part of the page. Page content is a separate optional request, disabled by default and limited to 2048 tokens per result unless you choose another supported budget. A length limit can leave relevant context outside the returned text.

Time filters need a date meaning as well as a window. published refers to publication time; crawled refers to the source crawl time. Octen’s Search reference excludes results missing the selected date field when filtering. Explicit start or end times take precedence over time_range. A recently crawled older article is still an older article.

Use domain and text conditions to focus the reading list. Octen’s Search reference gives domain exclusion precedence when the same domain is both included and excluded. Country sets a source priority rather than a language restriction. Narrow conditions can reduce the useful list; a requested count does not establish complete coverage.

Retain author, publication and crawl information when it is supplied, without replacing a missing date with the time of your request. Optional images are references associated with results; retrieving image files is separate. Summaries, translations, comparisons and factual checking are Agent work based on the material actually obtained.

Octen: Web Search sources and highlights

Octen: Search request reference

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

A highlight is selected text relevant to your query. Keep its source URL and quote only the passage actually supplied. If your conclusion depends on a wider section, request available page content or open the source separately. The Agent should distinguish retrieved wording from its own interpretation.

Octen Web Search for Source Pages

Octen Web Search for Source Pages

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Say what you want to find

Choose sources and how much to read

Check sources before sharing a summary

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Say what you want to find

Choose sources and how much to read

Check sources before sharing a summary

Questions about this tool.

Can I cite a highlight as if I had read the whole page?

Why might a time-filtered search omit a page?

What happens if a domain appears in both lists?

Does country choose the language of the results?