# PromptLinks: Link Building Agent Template for AI Search | FlyMy.AI > Clone PromptLinks: two research agents find publishers for a new PR article and existing articles where an editor could add your link, in one page with two CSV files. Page: https://flymy.ai/templates/promptlinks/ Agent template v1.1 Updated 2026-10-27 # PromptLinks Two research agents find where your brand can earn links - a new PR article or an existing article - and publish one page with both tables. - Publishers for a new PR article, with quoted prices - Existing articles Google and its AI Overview already cite - One page, two CSV files, every source dated [Clone this project](https://app.flymy.ai/projects/fleet%3A670f20dc-faf9-4024-8d73-f3875c546624) [Use from Claude / Codex](https://flymy.ai/templates/promptlinks/#start) By FlyMy.AI · templates/promptlinks v1.1 PromptLinks and its tools: Collaborator (call time ($0.000037 per second), table 1 only; placements are separate Collaborator purchases), DataForSEO (provider charge + 15% on the shared key: $0.0023-0.0046 per Google search with the AI Overview, measured), Tavily ($0.008 per basic web search (1 credit), measured), Two research agents (the lead starts both in parallel; each bills its own model and tools), Code execution ($0.0666 per sandbox hour, billed per second; about $0.001 per run, measured), Artifact page (free; one private page on app.flymy.ai per run, $0 measured), Chat files (free; both CSV files, the results and the page in the chat, $0 measured) ## Proof - **10 / 10** existing articles for 5 prompts on flymy.ai 2026-10-27 - **$0.10-0.15** a run of all three agents, 3-5 prompts, 4 runs 2026-10-27 - **1.5-3 min** from start to the page 2026-10-27 - **Frozen + SDK** frozen version and Python SDK runs passed 2026-10-27 - **Clone run** a copy with its two agents ran on its own 2026-10-27 - **[Public](https://app.flymy.ai/projects/fleet%3A670f20dc-faf9-4024-8d73-f3875c546624)** reference project to clone **3** agents lead + helpers **7** tools priced below **on demand** runs schedule after the first report **$0.10-0.15** per run measured ## What it does - Table 1: publishers for a new PR article from your Collaborator catalog, with the quoted price, format, writing and a named metric. - Table 2: published articles that rank or are cited in Google's AI Overview for your prompts, with the section, a quote, your page and an anchor. - One private page explains both workflows and carries both CSV files, the sources, observation times and what could not be checked. - Recommendations only: no outreach and no purchases; availability stays unknown until an editor answers. **Ideal for** Founders, PR and SEO teams who want their brand linked where search and AI answers already look. ### Connects to - **Collaborator** call time your account - **DataForSEO** ~$0.003 / search shared connection - **Tavily** $0.008 / search shared connection - **Two research agents** 2 agents built in - **Code execution** ~$0.001 / run built in - **Artifact page** free built in - **Chat files** free built in ## Project diagram from the reference project **The diagram as text** - **Inputs:** Website, brand and topic, 3-10 prompts, market and language, Research budget and tables - **Lead:** PromptLinks (claude-haiku-4-5) uses Two research agents, Code execution, Artifact page, Chat files - **Helper:** PromptLinks - publisher researcher (gpt-6-luna) uses Collaborator, Code execution, Chat files - **Helper:** PromptLinks - existing-article researcher (gpt-6-luna) uses DataForSEO, Tavily, Code execution, Chat files - **Tool:** Collaborator - call time ($0.000037 per second), table 1 only; placements are separate Collaborator purchases (your own account) - **Tool:** DataForSEO - provider charge + 15% on the shared key: $0.0023-0.0046 per Google search with the AI Overview, measured (shared FlyMy.AI connection) - **Tool:** Tavily - $0.008 per basic web search (1 credit), measured (shared FlyMy.AI connection) - **Tool:** Two research agents - the lead starts both in parallel; each bills its own model and tools (built into FlyMy.AI) - **Tool:** Code execution - $0.0666 per sandbox hour, billed per second; about $0.001 per run, measured (built into FlyMy.AI) - **Tool:** Artifact page - free; one private page on app.flymy.ai per run, $0 measured (built into FlyMy.AI) - **Tool:** Chat files - free; both CSV files, the results and the page in the chat, $0 measured (built into FlyMy.AI) - **Output:** Table 1: publishers for a new PR article (from PromptLinks - publisher researcher) - **Output:** Table 2: articles to ask for a link (from PromptLinks - existing-article researcher) - **Output:** One page with both tables and CSV files (from Artifact page) ## Day one from the reference project - 01 **Clone** The lead and its two research agents with their tested script. free - 02 **Connect Collaborator** For table 1: your customer account with the site-catalog API opened. Table 2 needs nothing. free - 03 **Answer once** Website, brand, topic, prompts, market, budget and tables in one message. free - 04 **First page** Both agents in parallel, then one page with two tables and two CSV files. $0.10-0.15 per run ## How it runs templates/promptlinks v1.1 Runs when you ask, in 1.5-3 minutes. Choose Schedule Agent under an answer to repeat it, for example every Monday at 09:00. ## The agent team from the reference project ### `PromptLinks` Lead lead Asks once, starts both research agents in parallel and publishes one page with both tables and both CSV files. Two research agents, Code execution, Artifact page, Chat files **Instruction excerpt** Two research agents of yours build two tables, and you publish both as one FlyMyAI artifact page with the two CSV files. Recommendations only: nothing is bought, published or sent to anyone. Budget: your research cap per run (default $1) on demand or weekly, claude-haiku-4-5, last run $0.144 on 2026-10-27 (flymy.ai, 5 prompts: 10 existing articles; table 1 unavailable, the test account's catalog is closed) ### `PromptLinks - publisher researcher` Table 1: new articles One page of your Collaborator catalog: up to 10 publishers with the quoted price, format, writing and Google AI Overview mentions. A closed catalog is reported with the fix. Collaborator, Code execution, Chat files **Instruction excerpt** Recommend publishers for a new PR or sponsored article: one page of the owner's own Collaborator catalog, at most 10 publishers, saved as new_article_placements.csv. Nothing is ordered, bought or published. Budget: half the research cap with the lead, gpt-6-luna, last run $0.003 on 2026-10-27 (catalog closed for the test account (HTTP 403), reported with the fix) ### `PromptLinks - existing-article researcher` Table 2: existing articles Google's top 10 and AI Overview references for your prompts plus web sources, then up to 10 articles read with robots.txt first: section, quote, your page and an anchor. DataForSEO, Tavily, Code execution, Chat files **Instruction excerpt** Find existing articles where the owner could ask the editor to add a link: Google results with the AI Overview references and web sources, then bounded public reads, at most 10 rows saved as existing_article_opportunities.csv. Budget: half the research cap with the lead, gpt-6-luna, last run $0.062 on 2026-10-27 (flymy.ai, 5 prompts: 10 articles, research data $0.056) ## You will be asked once, in one message - **Which website and brand?** Your site and the name you want cited. `example.com; Example` - **Which topic and prompts?** What the articles should cover and 3-10 searches your buyers make. `AI agent platform; best ai agent builder; run ai agents in the cloud` - **Which market and language?** The Google market and language (default United States, English). `United States; English` - **What budget and which tables?** Research cap per run (default $1), the highest price per new article, both tables or one. `$1; both` Keys and passwords never go into the chat: you connect accounts on their own pages. ## What it costs Measured on 2026-10-27 **0%** template markup or purchase fee [Usage pricing →](https://flymy.ai/pricing/) **Price of one run** | Item | Unit price | Per run | Cost | |---|---|---|---| | Google results with the AI Overview (DataForSEO) | provider charge + 15% on the shared key | 3-5 searches | $0.007-0.016 | | Web sources (Tavily) | $0.008 per basic search | 3-5 searches | $0.024-0.040 | | Collaborator catalog | call time, $0.000037 per second | 2 calls | under $0.001 | | Lead: Claude Haiku 4.5 and the run check | per token | about 6 model calls | ~$0.08 | | Two research agents: GPT-6 Luna | per token | about 8 model calls | ~$0.005 | | Code execution, page and files | $0.0666 per sandbox hour; page and files free | 3 calls, 1 page | ~$0.001 | | One run, 3-5 prompts, both tables | $0.10-0.15 | | | Measured 2026-10-27 on the reference and a clone: $0.144 (flymy.ai, 5 prompts), $0.115 (python.org, 3 prompts, frozen), $0.102 (docs.flymy.ai, 3 prompts, Python SDK) and $0.146 (a clone, 5 prompts). Table 1 stayed unavailable in these runs (the test account's Collaborator catalog is closed); its calls cost under $0.001. Every run shows its itemised price. ## What stays yours the same for every template ### Your connections stay yours A copy never carries the author's keys or connections. ### Nothing runs until you answer Clone creates the agents and a setup chat; nothing bills before your reply. ### A budget you set 5 / 20 / 50 USD per day / week / month by default; a run cannot outspend it. ## Versions a copy keeps the version it was made from - v1.1 2026-10-27 PromptLinks and CiteAI merged into one project: a lead with two research agents - new articles from your Collaborator catalog, existing articles from Google and its AI Overview - one page with two CSV files, a checksum-verified script hosted by FlyMy.AI. - v1.0 2026-10-21 First public brief: Collaborator publisher research, crawler checks and disclosed model assessments. ## How to start 3 agents, 7 tools, your budget 01 - Clone in FlyMy.AI Open the reference project and press Clone; a setup chat asks once. [Clone this project](https://app.flymy.ai/projects/fleet%3A670f20dc-faf9-4024-8d73-f3875c546624) 02 - Your Claude or Codex Connect FlyMy.AI with [skill.md](https://flymy.ai/skill.md) and paste the request below. 03 - API and SDK Agents MCP `projects_copy` with `fleet:670f20dc-faf9-4024-8d73-f3875c546624`, or `client.projects.copy`. $ SETUP REQUEST ``` Read https://flymy.ai/templates/promptlinks/template.txt and help me set up PromptLinks using my own connections. Ask for the missing inputs and budget before running it. ``` [Download brief](https://flymy.ai/templates/promptlinks/template.txt) ## FAQ **Is this a working project or just a brief?** A working, public reference project: a lead and its two research agents. Clone it and its first chat asks everything in one message; the page with both tables arrives in 1.5-3 minutes. **What is the difference between the two tables?** Table 1 lists publishers that sell a new sponsored or PR article, with the quoted price from your Collaborator catalog. Table 2 lists articles already published that rank on Google or are cited in its AI Overview for your prompts, with the section where your page fits: you ask the editor to add a link. **Why is table 1 unavailable?** Table 1 reads the site catalog of your own Collaborator customer account, which Collaborator opens for the API per account on request. Until it is open, table 1 comes back unavailable with that reason and what to do, and table 2 is still made. **Does it contact publishers or buy links?** No. It only recommends: it never writes to a publisher or an editor, never orders or buys a placement and never publishes an article. Availability stays unknown until an editor answers you. **Will this get my brand cited by Google or ChatGPT?** Nothing guarantees indexing, rankings or citations. The AI Overview references and Collaborator's AI Overview counts are Google observations at that time; the page does not measure what ChatGPT, Claude or Gemini cite. **What does a run cost?** Measured on 2026-10-27: $0.10-0.15 for a run with 3-5 prompts and both tables. Research data is about $0.04-0.06 of it, the rest the three agents' model tokens. Every run shows its itemised price. **Can it run every week?** Yes. After a good answer choose Schedule Agent under it with your day, time and time zone. The lead runs the frozen versions of its two agents; if FlyMy.AI asks to freeze them first, freeze each from its run in the chat, then schedule the lead. **Illustrative example, not a live result** ``` table 1 - publish a new PR article publisher: quoted price: , writing included: metric: Google AI Overview mentions (Collaborator), observed table 2 - add your link to an existing article article: "> section: - "" suggested link: , anchor: availability: unknown until the editor answers artifact page: ``` ## Sources [Reference project →](https://app.flymy.ai/projects/fleet%3A670f20dc-faf9-4024-8d73-f3875c546624), [Collaborator MCP →](https://flymy.ai/mcp/collaborator/), [DataForSEO MCP →](https://flymy.ai/mcp/dataforseo/), [SEO / GEO research template →](https://flymy.ai/templates/seo-geo/), [Usage pricing →](https://flymy.ai/pricing/) ## More templates Agent template ### [Instagram Trend Watcher](https://flymy.ai/templates/instagram-trend-watcher/) Finds the standout posts of the Instagram creators you pick and sends a daily digest with reuse ideas. Agent template ### [SEO / GEO Research Agent](https://flymy.ai/templates/seo-geo/) Crawls your site, checks your queries in Google and its AI Overview and publishes a report page with up to 10 actions, each with its evidence. Agent template ### [Website QA - desktop and mobile](https://flymy.ai/templates/website-qa/) Checks every link, opens each page in a real cloud browser on mobile and desktop, reads the screenshots for visible defects and publishes a report of what to fix. Content and connection guidance checked 2026-10-27. See linked documentation for current availability and prices. --- For agents: the FlyMyAI operating guide (sign up, connect, run models, tools and agents) is https://flymy.ai/skill.md, and https://flymy.ai/llms.txt lists the other pages and the documentation.