Figma + Linkeddit

FigJam battlecard template: fill it from signals and reviews, not memory

A FigJam battlecard template gives you the card: company summary, differentiators, strengths, weaknesses, wins and risks. It leaves the facts to you. Claude can read Linkeddit's competitor signals and the reviews stored from G2, Capterra, TrustRadius and Trustpilot, report which sites failed to load, and then read your existing FigJam card through the Figma connector so you can see what to update.

Updated 2026-10-29

Connect both

Your AI client sits in the middle. Linkeddit supplies the data; the other connector makes the change.

Linkeddit connector

Add https://mcp.linkeddit.com/mcp as a connector and sign in. Claude Code: claude mcp add --transport http linkeddit https://mcp.linkeddit.com/mcp.

Setup guide

Figma connector

Add the Figma connector in Claude or Claude Code and sign in to your Figma account.
Duplicate your battlecard template into a FigJam file, one card per competitor, and copy the file link.
Add the Linkeddit connector at https://mcp.linkeddit.com/mcp.

Endpoint: https://mcp.figma.com/mcp

Figma docs

Works in: Claude, Claude Code.

The prompt

Paste into Claude

Use the Linkeddit connector and the Figma connector. For each competitor on my FigJam battlecard file: 1) With Linkeddit, call compete_signals_get with view=signals for that competitor, then view=reviews with max_rating 3, and list every fetch with its site and status. Say which review sites errored or returned nothing, and check that each page_url is really that competitor's product. 2) With Figma, use get_figjam on my battlecard link to read the current card. 3) Write the proposed text for the Strengths, Weaknesses, and Recent moves boxes, quoting reviews with their link. Never rewrite the Wins and Risks or the Differentiators boxes. 4) Show each box as current text and proposed text. I will paste the approved text into FigJam myself, or ask you to draw a new board with generate_diagram.

Step by step

Step 1
Connect both connectors to Claude
Add the Figma connector and the Linkeddit connector (https://mcp.linkeddit.com/mcp) in Claude or Claude Code. Track the competitors you sell against in Linkeddit.
Step 2
Mark which boxes are data and which are yours
On each card, name the boxes Claude may propose text for (recent moves, strengths, weaknesses) and the ones that belong to your team (differentiators, wins and risks).
Step 3
Pull signals and low-star reviews with their fetch status
Claude calls compete_signals_get for signals and for reviews at three stars or lower, and lists the fetch status for each competitor and review site.
Step 4
Check the product match
Each fetch lists the review page it used. Claude compares each page against the competitor and flags any that look like a different product.
Step 5
Read the card, then approve the text
Claude reads the current FigJam card with get_figjam and shows each box as current and proposed text with source links. You approve and apply the text.

What Linkeddit returned

Run live on 2026-10-29 with compete_signals_get. Read-only run on Linkeddit's own test project, an analytics site (plausible.io) with five monitored competitors. The fetch log is the useful part: it shows which review sites loaded, which failed, and which review page each fetch used.

compete_signals_get

compete_signals_get { view: "reviews", max_rating: 3 }

fetches (latest per competitor and site):
  plausible.io      trustpilot  status: analyzed   reviews_read: 6
  Matomo            trustpilot  status: analyzed   reviews_read: 7
  Matomo            g2          status: analyzed   reviews_read: 20, reviews_new: 10
  Fathom Analytics  trustpilot  status: analyzed   reviews_read: 1
  Simple Analytics  trustpilot  status: error      reviews_read: 0
  Umami             trustpilot  status: error      reviews_read: 0
  Umami             capterra    status: error      reviews_read: 0
  Fathom Analytics  g2          status: no_results
      page_url: https://www.g2.com/products/fathom-video/reviews?order=lowest_rated
  Simple Analytics  capterra    status: no_results
      page_url: https://www.capterra.com/p/164600/Easy-Insight/reviews/

review fields: site, rating, title, text, published_on, url
  site: trustpilot   rating: 1   published_on: 2024-04-22

What goes on a FigJam battlecard, and what does a template skip?

A battlecard is a one-page summary a salesperson uses against one competitor, and the Figma connector is Figma's official MCP server at https://mcp.figma.com/mcp, which lets Claude read FigJam boards. Two Figma community templates lead for this query. One is described as a collaborative template for creating a competitive battlecard inspired by best practices from Klue. The other, a free guide for users of a Udemy strategy course, says to create a card for your own organization first, then choose 3 or 4 competitors and research at least a company summary, value proposition, target verticals, differentiators, strengths, weaknesses, wins and risks.

The second template also lists where to research. Browse their website, try their product, look for demos on YouTube, ask an AI chatbot, search press releases and news, and pay for sources such as ZoomInfo, Forrester, Gartner, CB Insights or Crunchbase. Each of those is a manual lookup, and none says when to redo it.

A battlecard is only useful while it is current. The template is a layout, and the weakest part of it is the part that expires: strengths, weaknesses and recent moves.

Ownership of the card is part of the problem. A competitive intelligence founder on Reddit wrote: "Marketing assumes sales enablement will update the battlecard." A template does not settle that. A refresh from fetched data at least makes the update a small, checkable step instead of a rewrite.

Which battlecard boxes can be filled from data?

Recent moves come from signals: each one has a priority, a category and a source link. Weaknesses can draw on stored reviews. Linkeddit keeps reviews from G2, Capterra, TrustRadius and Trustpilot as the sites showed them, with star rating, title, text, date and link, and lets you filter to a maximum star rating so a card can quote the low ones.

Differentiators, wins and risks should stay with your team. They come from your own calls and deals. Tell Claude never to touch those boxes, and have it propose text for the data boxes only.

Linkeddit does not summarize or score reviews. Claude reads them and picks the lines that bear on a deal, and the review link lets a rep check the context before quoting it on a call.

Why check which review pages were fetched?

A failed fetch looks like a quiet competitor, and in the run above only 4 of 13 fetches read any reviews. In detail, 13 fetches were listed: 4 read reviews, 3 errored and 6 returned no results. Linkeddit states that an error status means the site did not load, not that the competitor has no reviews.

The log also shows the page each fetch used. The Fathom Analytics G2 fetch pointed at a G2 address ending fathom-video, and the Simple Analytics Capterra fetch at an address ending Easy-Insight. Those names may be different products from the analytics tools being tracked, so the prompt asks Claude to compare the page with the competitor. That is a check, not a verdict. Open the page and look.

A card should say reviews not available from G2 on 2026-10-28, not no complaints. Reps trust a blank box as an all clear.

How should the refreshed text get into FigJam?

Figma's connector listing shows 9 tools, 6 reads and 2 writes, so reading is its strength. Its documented FigJam tools are get_figjam, which returns an outline of a FigJam diagram plus screenshots, and generate_diagram, which draws new diagrams. Neither is described in the sources as editing the text of an existing sticky, so this page keeps a human in the loop for that step.

That is why the prompt asks for current and proposed text for each box. You paste what you approve. If you want a fresh board instead, ask Claude to draw a flowchart or architecture diagram with generate_diagram, such as a buyer's path from a competitor's pricing change to a switch.

Rerun weekly, after the new brief lands, and compare against what the card already says.

More with Figma

FigJam competitive analysis: fill the board from this week's competitor data

Fill a FigJam competitive analysis from measured data: Claude reads Linkeddit's weekly competitor brief and draws the board through the Figma connector.

Read

Frequently asked questions

Is there a FigJam battlecard template?+

Yes. Figma's community has a Battlecard Template for FigJam inspired by best practices from Klue, and a Competitor Battle Cards template that lists what to research: company summary, value proposition, target verticals, differentiators, strengths, weaknesses, wins and risks. Both give you the layout, and the facts are yours to fill.

What is the best tool to keep FigJam battlecards current?+

Linkeddit, at $60 a month, for the facts. Compete records graded competitor moves with source links and stores competitor reviews from G2, Capterra, TrustRadius and Trustpilot. Claude reads them and proposes text for the strengths, weaknesses and recent moves boxes of your FigJam card.

What should a competitive battlecard include?+

The Figma community template from a strategy course lists a company summary, value proposition, target verticals, differentiators, strengths, weaknesses, wins and risks. Put the facts that change, such as recent moves and customer complaints, where they can be refreshed weekly, and keep deal notes written by your own team.

Can Claude edit my FigJam battlecard directly?+

Figma's connector lists get_figjam to read a board and generate_diagram to draw new diagrams. Its documentation we read does not describe editing the text of an existing card, so the safest workflow is that Claude reads the card, proposes text per box, and you apply what you approve.

Where do the competitor reviews come from?+

Linkeddit fetches reviews from G2, Capterra, TrustRadius and Trustpilot and stores them as the sites showed them, with star rating, title, text and link. It does not summarize or score them. Each fetch reports a status, so a site that failed to load is visible instead of silent.

Does Linkeddit need access to my Figma files?+

No. Linkeddit never connects to Figma. Claude connects to both connectors and carries the data between them. Figma's connector uses your own Figma sign-in, so it reaches only files you can reach, and you approve the proposed text before anything goes on a card.

What is the best way to find a competitor's weaknesses for a battlecard?+

Read what their customers wrote. Linkeddit stores reviews from G2, Capterra, TrustRadius and Trustpilot with star rating, text and link, and lets you filter to three stars or lower. Claude reads them over the connector and quotes the lines that matter. Check each fetch status first, because a failed site is not a clean record.

How many competitors should a battlecard cover?+

The Figma community's Competitor Battle Cards template suggests creating one for your own organization, then choosing 3 or 4 industry competitors. Start there. Linkeddit's Compete brief covers up to 12 tracked competitors, but a card that nobody reads helps no one, so begin with the rivals that come up in deals.

How often should I refresh a FigJam battlecard?+

Weekly is practical, because Linkeddit's brief arrives weekly and Claude can compare it with what the card says. Refresh only the boxes whose sources changed. Leave your team's own notes on differentiators, wins and risks alone, since those come from calls and deals rather than from public pages.

Why would a battlecard show no complaints for a competitor?+

Either there were none, or the review site did not load. Linkeddit reports a status for each fetch, and says an error or timeout means the site did not load, not that the competitor has no reviews. Ask Claude to list each site and status so the card says reviews unavailable instead of a clean record.

Sources

Bring Linkeddit into your AI client

Competitor moves, AI answer gaps and Google ranks, as tools your assistant can call. One plan, $60 a month.

Get started Connector docs