Keenable

View saved version →

---
name: keenable-deep-research
description: "Run comprehensive multi-source web research with Keenable and synthesize a cited answer. Use for literature reviews, market and competitor analyses, and detailed comparisons."
---

# Keenable Deep Research

Produce a thorough, well-cited answer to a broad question by iterating over the
Keenable search and extract tools. Use this for tasks that a single search can't
cover: literature reviews, market and competitor analyses, and detailed
comparisons.

## Workflow

1. **Decompose.** Break the question into 3-6 concrete sub-questions. State them
   before searching so the plan is auditable.
2. **Search per sub-question.** Run a focused `search_web_pages` query for each.
   Keep queries specific; split further when coverage is thin.
3. **Read the strongest sources.** Use `fetch_page_content` on the most relevant
   results to read them in full rather than trusting snippets for anything
   consequential.
4. **Cross-check.** For any claim that affects the conclusion, confirm it against
   at least one independent source. Note disagreements explicitly.
5. **Synthesize.** Write the answer in your own words, grouped by sub-question or
   theme. Cite every factual claim with its source URL. Separate sourced facts
   from your own analysis, and call out gaps or low-confidence areas.

## Guidance

- Track which sub-questions are answered and which still need work; keep going
  until the question is covered, not just until the first search returns.
- Prefer primary and authoritative sources; treat aggregators and undated pages
  with caution.
- For relative dates, compute exact dates from today before searching.
- The tools are rate-limited. If a long research run hits the limit, report what
  was covered so far and stop instead of retrying in a loop; never ask the user
  to paste an API key or any other credential into the conversation.

View saved version →

---
name: keenable-web
description: "Search and read the web with Keenable. Use for current information, source discovery, and extracting clean content from known URLs. Keyless by default."
---

# Keenable Web Research

Use the Keenable MCP tools provided by this plugin to retrieve current,
source-grounded web information. The tools are keyless by default — no
authentication is needed to call them.

## Choose the right tool

- `search_web_pages`: Find relevant sources when the user needs current or
  citable information from the web.
- `fetch_page_content`: Read the full, clean markdown content of a known URL.

## Workflow

1. Clarify scope, freshness, and output format when the request is ambiguous.
2. Use `search_web_pages` for focused questions and source discovery.
3. Use `fetch_page_content` when the user supplies a URL, or to read a promising
   result from search in full before relying on it.
4. Synthesize the results rather than pasting raw tool output. Preserve source
   URLs and cite factual claims.

## Search guidance

- Write concise, specific queries instead of long instructions.
- Split multi-part questions into distinct searches when that improves coverage.
- For relative dates such as "last week," calculate exact dates from today.
- Prefer primary and authoritative sources for factual or technical claims.
- Cross-check consequential claims against more than one independent source.
- Distinguish sourced facts from your own analysis.

## Limits

The tools are rate-limited. If a call reports that the limit was reached, say so
plainly and stop instead of retrying in a loop; usage limits are tied to the
connected Keenable account. Never ask the user to paste an API key or any other
credential into the conversation.