Promptwatch vs Outserp vs Agent Number 10 vs Outcited (2026)
Outserp, Agent Number 10 (formerly CiteAgent), and Outcited all start at $49 and promise agents that act. How they compare with Promptwatch in 2026.
All three newcomers start at $49 a month and pitch themselves as agents rather than trackers. Outserp writes and publishes articles at volume. Agent Number 10, called CiteAgent until October 2026, writes individual fixes and waits for your approval. Outcited runs a weekly autopilot that tracks, writes, publishes, and rewrites. For a founder or a small marketing team that wants AI search handled without hiring, each one looks like the answer.
They are worth taking seriously, and in places they come closer to a full loop than older agentic tools. But $49 buys a specific slice of the work, and the slice is different for each vendor. We compare them with Promptwatch because it is the platform the three are effectively competing with: one workspace that tracks AI answers, explains them, publishes content, and measures what that content earned.
The frame is our usual loop. Monitor: what does the tool see in AI answers? Diagnose: can it tell you why you lost? Act: does it change the site or ship content? Verify: can it show the change paid off? Facts for Outserp, Agent Number 10, and Outcited come only from their directory entries, read from vendor pricing pages and llms.txt files in October 2026. Anything an entry does not cover is marked "not published."
Promptwatch
Promptwatch is an AI search visibility platform built around the full loop. It monitors ChatGPT, Gemini, Claude, Perplexity, Grok, Llama, DeepSeek, Mistral, Copilot, and Google AI Overviews and AI Mode through the real user interfaces, not only APIs. Paid plans refresh prompts daily, and prompts carry search volumes, difficulty scores, and query fan-outs. Country tracking comes on every plan, and state and city tracking start on Professional.
For diagnosis, citation analytics shows the pages, domains, Reddit threads, and YouTube videos each engine cites, and citation trends show how that mix shifts over time. Agent Analytics logs AI crawler visits from ChatGPTBot, ClaudeBot, PerplexityBot, GoogleOther, and the Meta AI crawler, with a crawl-to-citation path and error tracking.
To act, Content Agents plan from a content gap analysis, write, and publish to Webflow, Framer, or WordPress through an official plugin. Every draft passes through a review inbox first. On Framer, going live redeploys the whole site. To verify, visitor analytics uses a lightweight script or GTM template to tie AI-referred sessions to conversions. There is also an MCP server and REST API v2 on all paid brand plans.
Explore is free with 10 ChatGPT prompts. Essential is $95 a month with a 7-day trial, 50 prompts, and 5 AEO articles, without crawler logs. Professional is $245 a month with 150 prompts, 25 million crawler logs, and 15 articles.
Outserp
Outserp is a London-registered AI SEO and AEO platform that puts a content agent first. Its 13-stage pipeline researches, writes, cites, optimizes, and publishes articles in your brand voice to WordPress, Ghost, Webflow, Sanity, or Shopify. A Content Grid handles bulk generation, and programmatic SEO templates and a workflow builder sit alongside it. A content audit scores pages on nine AEO dimensions with one-click fixes, and it can draft social posts of cited content for LinkedIn, X, Reddit, and YouTube.
Tracking covers mentions, citations, sentiment, and share of voice on ChatGPT, Perplexity, Claude, Gemini, Grok, and Google AI Overviews. An AI traffic view captures crawler and referral visits and pulls in GA4 AI-assistant sessions. Developer access is generous: a 95-tool MCP server, REST API, and CLI come on every plan, including the trial.
Starter is $49 a month ($34 annual) with 10 articles, 50 prompts, and one project. Growth is $149 ($87 annual) with 50 articles and 200 prompts. Scale is $399 ($279 annual) with 200 articles and 300 prompts. Enterprise starts at $1,999 a month. Every plan has a 7-day trial with no card. The prompt count stays small next to the article volume, and the pages do not say whether answers are read from live chat interfaces or APIs. Its own llms.txt says it does not cover local SEO, hreflang, or ecommerce product feeds. Visitor conversion analytics are not published.
Agent Number 10
Agent Number 10 is built by Akoota Labs and ran as CiteAgent until October 2026. Both domains still serve it, so older links and reviews use the old name. It checks whether ChatGPT, Gemini, Claude, Perplexity, and Google AI Overviews recommend your site, shows who gets recommended instead and which sources they use, then writes a fix. A fix is one finished change, such as an answer-led paragraph, an FAQ block, schema, a heading, or technical SEO work. You approve it before it ships, and the team either ships it for you or hands it over ready to paste.
It also tracks Google rankings and runs a site audit with keyword research and Search Console data, so it pitches itself as a replacement for an SEO tool plus an AI tracker. You can re-check the same questions on demand to see what moved. There is an MCP server for Claude, ChatGPT, Claude Code, and Cursor with OAuth sign-in.
Agent is $49 per domain a month with five approved fixes, 25 tracked buyer questions, 50 keywords, and a 500-page audit. Agent Pro is $129 per domain with 25 fixes, 50 questions, and backlink and off-page data. Agencies is $399 a month for ten client domains with 50 shared fixes. Billing is month to month with a 14-day refund on the first payment. Fix counts are hard caps, and the pricing page does not state how often AI answers refresh. The listing does not publish crawler logs, visitor conversions, or a content agent that publishes to a CMS.
Outcited
Outcited is the most complete loop of the three at its price. Each week it tracks buyer-style queries on ChatGPT, Claude, and Gemini, writes articles for the gaps, publishes them, and rewrites any article that has not been cited after three weeks. Articles go to WordPress, Webflow, Ghost, or Shopify, open as a GitHub pull request for a Next.js blog, go out by webhook, or download as MDX.
It also does something rare at $49. A small script logs AI crawler visits by bot (GPTBot, ChatGPT-User, ClaudeBot, and GoogleOther) and by page, and the dashboard lines up publish, crawl, and citation dates. That is a real verify step, even if it is lighter than a crawl-to-citation log with error tracking.
Launch is $49 a month for one domain, 10 queries, one competitor, and 10 articles. Growth is $79 with 100 queries, five competitors, 30 articles, and the rewrite engine. Scale is $149 for up to ten domains with 200 queries and white-label reports. All plans have a 7-day trial. The trade-offs are clear in the listing: weekly rather than daily tracking, three engines (no Perplexity, AI Overviews, or Copilot), answers that appear to come from APIs, and English only. Visitor conversion analytics are not published, and neither is Google rank tracking.
Side by side
| Promptwatch | Outserp | Agent Number 10 | Outcited | |
|---|---|---|---|---|
| Category | AI visibility platform | Agentic SEO automation | Agentic SEO automation | Agentic SEO automation |
| Entry price | Free Explore; Essential $95/mo | $49/mo ($34 annual) | $49 per domain/mo | $49/mo |
| Trial | 7 days on Essential | 7 days, no card | 14-day refund on first payment | 7 days |
| Monitor | Daily prompts on 9 LLMs plus AI Overviews and AI Mode | 50 to 300 prompts on 6 engines | 25 to 50 buyer questions on 5 engines; refresh not stated | Weekly, 10 to 200 queries on 3 engines |
| Diagnose | Citation analytics; Agent Analytics crawler logs from Professional | 9-dimension AEO audit; sentiment and share of voice | Who gets recommended instead and their sources | Per-engine breakdown, competitor tracking, bot crawl script |
| Act | Content Agents publish to WordPress, Webflow, Framer with review inbox | 13-stage content agent publishing to 5 platforms | Approved fixes shipped by team or ready to paste | Articles auto-published; rewrites after 3 weeks uncited |
| Verify | Visitor analytics with conversions; prompt and citation trends | AI traffic view with GA4 AI-assistant sessions | On-demand re-checks | Publish, crawl, and citation timeline |
| MCP or API | MCP server, REST API v2 | 95-tool MCP, REST API, CLI | MCP server | Not published |
Who should pick which
Pick Outserp if your bottleneck is output: you know what to write, you want dozens of articles a month, and you value an API you can script against. Pick Agent Number 10 if you have a small site, no writer, and want a handful of finished, approved fixes each month for one flat per-domain price. Pick Outcited if you run an English-language site, can live with weekly data on three engines, and want the cheapest hands-off loop that still shows when AI bots crawled a page.
Pick Promptwatch if the program has to stand up to scrutiny. The newcomers either track a few dozen prompts or check them weekly, and none of them publishes visitor analytics that tie an AI-referred visit to a conversion. When someone asks whether the articles are earning revenue, those are the numbers you need.
Our recommendation
Evaluate Promptwatch first, then decide whether a $49 agent adds anything on top. Promptwatch already runs the act step these tools sell, through Content Agents that plan from content gaps and publish to WordPress, Webflow, or Framer behind a review inbox. It then covers what they leave thin: citation analytics down to Reddit and YouTube, Agent Analytics crawler logs with a crawl-to-citation path and error tracking, and visitor analytics with conversions. Start on Professional at $245 a month to get the crawler logs, or open a free Explore account at Promptwatch to see where you stand today.