Highspot MCP with Claude: check your sales content against competitor moves and AI answers

Highspot is where approved sales content lives. Its Agent Index listing shows 7 tools, 6 reads and 1 write: search_content, get_item_content, get_content_answer, get_content_recommendations, lookup_deal, get_deal_answer and generate_pitch. The Claude directory describes the connector as a way to search for content, get answers, see analytics and take action, and says it is part of Seismic. It knows your own content and deals. It has no view of what a competitor published yesterday.

Linkeddit is the outside view. Compete watches up to 12 competitors weekly and returns each move with a priority, category, source link and date, with reviews from G2, Capterra, TrustRadius and Trustpilot stored separately. Answer Radar asks ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews your buying questions and records who is named, recommended and cited. Both come with one plan at $60 a month.

Claude is the analyst between them. A competitive intelligence founder on Reddit described why battlecards rot: "Marketing assumes sales enablement will update the battlecard." The workflow here gives that update a trigger and an evidence trail, and the person who owns the card still approves the change.

Add Highspot from Claude's connector directory and sign in with your Highspot credentials. Microsoft Learn's page for the Highspot connector in Microsoft 365 Copilot says you must be a licensed Highspot user, and that users sign in with their Highspot credentials to get the tokens the connector uses. The same page lists a limit of 100 API calls per connection every 60 seconds for that connector, so keep a first run small.

The Agent Index lists the connector's permissions as read and write, but only generate_pitch is classified as a write. The other six read content and deal information. That makes this a low risk connector to start with: Claude can find and read your content, and nothing in the tool list says it can edit a battlecard in place.

Then add the Linkeddit connector at https://mcp.linkeddit.com/mcp. The two connectors are independent. Highspot never sees Linkeddit's data, and Linkeddit never sees your content.

Have Claude read the card, read the evidence, and list the lines that no longer hold. It finds the card with search_content, reads it with get_item_content, then asks Linkeddit for the competitor's recent moves and its stored reviews, including low-rated ones through the max_rating filter. It returns a short table: each card line, what the evidence now says, a source link and a date.

On Linkeddit's test project, the latest week held 2 high priority moves: Fathom launching its Luna AI assistant and Matomo publishing critical PHP security guidance, each with a source link and a reason it matters. A card that said a competitor had no AI assistant would be wrong the day the Fathom changelog went up, and Claude can flag exactly that line.

Highspot's tool list has no step that edits a content item, so Claude drafts the replacement lines and you or the card's owner paste them into Highspot. That is a feature for this job: the person who owns the card sees every changed line and its source before a seller does.

Ask Claude to read the latest Answer Radar results from Linkeddit and write a one-page summary. On the test project, plausible.io, the latest run measured five of six engines: presence was 8 of 9 answers on ChatGPT, 8 of 9 on Gemini, 7 of 9 on Google AI Mode, 6 of 6 on Google AI Overviews and 5 of 9 on Claude. Perplexity returned errors on all 10 samples and came back unmeasured, not zero.

Per engine, Linkeddit also returns share of voice, mean position, sentiment and the domains cited. In the Claude result, Plausible held 17 percent of vendor visibility, with Fathom and Matomo at 12 percent each. The engines where a competitor is named instead of you are the lines for sellers, because a buyer may already have seen that answer before the call.

Highspot's connector can search and read content, not publish a report, so the summary goes to the team through your usual channel. Claude can use search_content to check whether any existing Highspot asset already answers the same buying question.

Filter to high priority, then act on those. A rule that alerts sellers only on high priority moves would have sent 2 alerts in the test project's latest week and kept the rest in Linkeddit, which is the volume a seller can read. Each alert needs the move, the source link, the date and one line on why it matters to a deal.

Claude can add the deal context. With lookup_deal and get_deal_answer it can read what Highspot knows about a named deal, and with get_content_recommendations it can see what content Highspot suggests for it. Ask Claude to say which deals mention the competitor, then check the match yourself. A competitor name in a deal is not proof the buyer is comparing you.

Highspot's tool list has no alert or notification tool. Claude writes the alert and you send it, or paste it where your team already reads. Linkeddit's own monitors run weekly, so the alert is ready when you ask, not the minute the page changes.

When an engine recommends a competitor, it usually cites pages. Linkeddit lists your open gaps and the sources cited behind each one. On the test project it showed 1 open gap, "Umami vs Matomo for self-hosted analytics", with an evidence score of 22, and the run's citation share named domains such as analytics-alternatives.com and bootstrap.build.

Ask Claude to take that buying question and search Highspot with search_content to see whether your enablement content answers it, then read the best match with get_item_content. If nothing answers it, you have a gap a seller will hit, and Claude can draft the missing section for review.

This is a check, not an outreach list. Linkeddit shows what the engines cited and Claude shows what your content lacks. The decision about what to publish stays with your team.

Claude reads through your own Highspot sign-in, so it can reach what your account can reach. Per the Microsoft Learn page, Highspot access needs a licensed user. Use content you are comfortable having Claude read, and name the card or collection in the prompt rather than asking it to search everything.

Then read the output. Every flagged line should carry a source link and a date, and the report should say which parts are measured. On the test project, Linkeddit's weekly brief returned an error and the competitor brief for Fathom came back empty, and a good report says that instead of filling the space.

Count the changes made, not the reports written. Track how many card lines Claude flagged, how many the owner confirmed as stale and changed, and how many days the card ran with a known error. In the test project's latest run, five of six engines were measurable, so also watch the engines where a competitor is named instead of you, week over week.

Common mistakes are letting Claude rewrite a card nobody reviews, asking it to search the whole library on the first run, and treating one run as a trend. That run took one sample per question, so look for a gap that repeats before you change a sales message.