MCP
Refresh battle cards from rivals' launch videos
Import each launch, summarize claims, and query the differences — battle cards grounded in what they actually said.
library_import_youtube_videorun_youtube_tooldeep_contextThe problem
Battle cards age like milk. The competitor section was accurate the sprint someone wrote it; since then the rival has launched twice, repriced once and quietly renamed the feature your card attacks. Your reps discover the drift live, mid-call, when the prospect corrects them — which converts your battle card from a weapon into a liability.
The refresh material publishes itself: every rival launch comes with a video — the demo, the keynote segment, the feature walkthrough — containing the exact claims your reps will hear quoted back. Keeping cards current was never an information problem; it was a processing one.
How veridive MCP helps
library_import_youtube_video captures each rival launch as it drops — tagged by competitor, transcripts and all, building the claims archive battle cards should be written from. run_youtube_tool's summarizer and key-quotes process each launch fast: what shipped, what was claimed, the exact sentences marketing chose.
deep_context does the comparative layer battle cards actually need: "what does their new launch claim versus their last one? which of our counter-points did this launch neutralize?" — cited to the launch videos. The card refresh becomes a post-launch ritual measured in prompts: their event ends, your cards update the same week, and reps stop being corrected by prospects.
How to run it
- 1Connect veridive MCP
One-time setup: add the veridive MCP server to Claude, ChatGPT, Cursor or any MCP client and generate your API key. Takes about two minutes.
- 2Capture the launch
The launch video, keynote segment and demo — imported and tagged the day they publish.
Import these <competitor> launch videos into my library, tagged '<competitor>-launches': <video URLs> — use veridive's library_import_youtube_video. - 3Process the claims
Summary plus verbatim claims per video — the raw material for the card.
Summarize each imported launch video and pull the key claims verbatim — features, numbers, positioning statements. - 4Refresh the card comparatively
What changed, what got neutralized, where the new soft spots are — asked across the launch archive.
Comparing this launch to their previous ones in my library: what's genuinely new, which of our existing counter-points still hold, and what new gaps did they expose? Cited.
Problems it solves
- Battle cards drift out of date between writes; reps get corrected by prospects mid-call.
- Launch-claim archaeology across a rival's videos takes hours nobody schedules.
- Cards state features without the rival's exact phrasing — the phrasing reps will hear quoted.
- 'What changed since last launch' requires a memory nobody has and an archive nobody kept.
Why it's valuable
- Cards refresh the week of the rival's launch — the drift window closes.
- Claims arrive verbatim and cited; counter-positioning attacks what was actually said.
- The launch archive compounds: every event makes the comparative questions sharper.
- Enablement's refresh cost drops to prompts, so it actually happens on cadence.
Example prompts
Import <competitor>'s launch videos, tagged: <URLs> — use veridive's library_import_youtube_video.Summaries + verbatim key claims per launch video.Vs their previous launches: what's new, what holds, what gaps opened? Cited.FAQ
How does this connect to the demo win/loss use case?
Same enablement stack, different layers: demo-mining studies the rival's ongoing sales performance; this one processes launch events into card refreshes. Launches set the claims, demos show the delivery — cards need both.
What if a rival launches without a video?
Rare and shrinking — but then the sweep documents the absence, and your card notes the claims exist only in text. The video layer is where demos over-promise, which is exactly what cards weaponize.
Who owns this ritual?
PMM typically — it's a per-launch, prompt-driven task that slots after any rival event. The library tags keep multiple competitors' archives separate and queryable.
Related use cases
Watch how rivals demo the features you lose deals on and get a cited summary of the exact claims they make.
Compare how a rival described their product a year ago vs today; answers cite the exact moments the pitch changed.
Import every competitor's channel and interrogate all of it in one conversation — "what did they emphasize this quarter that they didn't last?"
Run this in your own assistant
Connect the veridive MCP server to Claude, ChatGPT or Cursor and paste the prompts above — setup takes about two minutes.