agentsclimarketplace

Rapport building

Skill sumitaich1998/OpenDate/src/opendate/skills/registry/rapport-building

Open-source AI dating agent — bring your own Tinder token and any LLM (OpenAI, Claude, Gemini, DeepSeek, Qwen, and more). Screens dates, makes matches, and chats in your own voice, with consent-first guardrails.

Install
npx -y skills add sumitaich1998/OpenDate --skill rapport-building

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Finds common ground, asks great open questions, listens actively, and reflects back — turning small talk into genuine connection.

SKILL.md

2.3 KB, as published. Nobody here has run it

Rapport building

Chemistry needs jokes; connection needs rapport. This is where you turn surface chat into the feeling of "oh, I actually like talking to this person."

The loop

Ask → listen → reflect → relate → (gently) deepen.

  1. Ask an open, specific question about something they care about.
  2. Listen for what they emphasize or get animated about — that's the thread.
  3. Reflect it back so they feel heard ("so the move was leaving the safe job — that's a big swing").
  4. Relate with something true about yourself. Reciprocity builds trust.
  5. Deepen by one layer: from what → why → what it meant.

Great questions

  • Trade "what do you do?" for "what's something you're weirdly good at?"
  • Trade "do you travel?" for "what's a place that changed how you think?"
  • Ask about meaning and motivation, not résumés: "what got you into that?", "what part do you actually love?", "what's lighting you up lately?"
  • Follow energy, not your checklist. If they light up about their dog, stay there a while.

Active listening on text

  • Name the feeling, not just the fact: "that sounds like it was equal parts terrifying and freeing."
  • Callback to something they said earlier to show you retained it.
  • Match self-disclosure depth. Go one layer deeper than they did, never ten.
  • Validate before pivoting. Acknowledge what they shared before adding yours.

Calibration

  • Keep it warm and curious, not clinical. One good question per message.
  • Balance depth with lightness so it never feels like therapy. A little humor keeps it human.
  • Watch the give/take ratio — aim for roughly 50/50 sharing.

Do / Don't

  • Do reflect feelings and motivations, not just facts.
  • Do share something real about yourself in return.
  • Don't rapid-fire questions; it feels like a form.
  • Don't one-up their stories — relate, then hand it back.

Output

A warm message that makes them feel heard, shares something true in return, and opens one genuine door to go a layer deeper.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.