agentleFS
oncesylviaoncesylvia/fundraising-skills/skills/warm-path-finder/SKILL.md

warm-path-finder

Skill4 starsChanged 5 months agoMITView on GitHub

Find the right human at an investor (the partner / principal / senior investment manager who owns your deal) AND map the warm-connection paths to reach them. Use when a founder says "who at <firm> should I talk to", "how do I get a warm intro to <investor>", "find a connection to <fund>", "who can introduce me", or is stuck going from a target firm to an actual person and a path in. Works from the founder's own authorized data (Gmail, exported LinkedIn connections, contacts) + public sources — never scrapes private data or guesses personal emails.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read shared/references/outreach-ethics.md first.

    line 19
  2. Distinctive:

    If a step would cross this line, stop and tell the founder.

    line 33
  3. Distinctive:

    Always note the contact route + DM-open/submit status you found, and a reminder to re-verify it right before reaching out.

    line 110
  4. Very specific:

    If the founder is raising in the China market, read shared/references/china-market-playbook.md and use references/zh-talk-tracks.md.

    line 125
  5. Distinctive:

    do not: ask a China founder for a LinkedIn CSV.

    line 128

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Warm path finder

Targeting tells a founder which firms. This skill closes the gap everyone gets stuck on: which specific human to reach, and who can warmly introduce them. It does two jobs:

  1. Pinpoint the person — the partner / principal / senior investment manager who actually owns your sector and stage at the target firm (pitching the wrong person wastes the shot). See references/finding-the-person.md.
  2. Map & rank warm paths to that person — surface every legitimate route from the founder's network, score by trust × reachability, and hand the best one to the warm-intro skill. See references/connection-mapping.md.

Read shared/references/outreach-ethics.md first. The whole skill lives or dies on staying inside what's authorized and public (see the data-source tiers below). Same anti-hallucination rule as the other research skills: a person or a connection you assert must come from real data or live research, with a source — never invent a partner, a title, or a mutual contact.

The hard line on data (read this before doing anything)

✅ Allowed❌ Not allowed
The founder's own authorized data: their Gmail, their Google/phone contacts, a LinkedIn connections export they downloadedReading the founder's LinkedIn graph via API or scraping (no API exists; scraping breaks ToS and risks a ban)
Public info: X/Twitter bios, firm team pages, portfolio lists, podcasts, conference speakers, Crunchbase/AngelList peopleCrawling login-walled content at scale
The founder manually checking shared connections / DM-open status while logged inGuessing or buying personal email addresses to blast

If a step would cross this line, stop and tell the founder.

Step 1 — Pinpoint the right person

For a target firm, identify the specific decision-maker, not "the firm." See references/finding-the-person.md. In short:

  • Find the partner/principal who leads your sector & stage — via the firm's team page, who led the round for portfolio companies like yours (search "<analogous company> <round> <firm> partner"), their public thesis/posts.
  • A senior associate / principal / senior investment manager who covers your space can be the better first target than a famous GP — they're more reachable and looking to source. Note who sources vs. who decides.
  • Capture their public footprint: firm bio, LinkedIn URL, X handle, recent writing/podcasts. This is research input, not contact-harvesting.

Step 2 — Search the founder's own authorized data first (Tier 1)

This is where the real warm paths hide, and it's 100% authorized. Ask the founder which they can provide, then mine it. If they don't know how to get the data, point them to references/exporting-your-data.md (step-by-step LinkedIn export + Gmail/contacts, EN + 中文).

  • Gmail (if connected): search their history for anyone already linked to the target firm or person — past threads, shared recipients, intros they've received. An existing thread is a warm path.
  • Exported LinkedIn connections CSV: have the founder export it themselves (LinkedIn → Settings → Data Privacy → Get a copy of your data → Connections). Then cross-reference names/companies against the target firm and its portfolio.
  • Contacts: Google/phone contacts for the same matching.

Match on: people at the firm, people who are portfolio founders of the firm (highest-trust introducers), and people who share an employer/school with the target person.

Step 3 — Map public-network paths (Tier 2)

For paths not in the founder's own data, find them in the open:

  • Portfolio-founder paths (best): a founder the firm already backed vouching for you is the strongest intro. Find portfolio cos near your space; the founder may know one, or know someone who does.
  • X/Twitter: locate the firm's partners/managers, their public thesis, and whether they're reachable (open DMs / active repliers). The founder confirms DM-open status on the profile; you draft the message.
  • Shared communities: accelerator alumni, university/employer networks, angel syndicates, founder Slacks/Discords, mutual advisors.
  • Events: who from the firm is speaking where — a live intro beats a cold DM.

See references/connection-mapping.md for how to mine each surface.

Step 4 — Score and rank the paths

For every path found, score it so the founder spends their best shots well. Rank by trust × reachability (see the rubric in connection-mapping.md):

  • Trust: how much does the target person trust the introducer? (portfolio founder > close peer/colleague > acquaintance > weak tie)
  • Reachability: how easily can the founder actually ask this introducer, and how likely are they to say yes and follow through?

Output a ranked shortlist, not a dump. The best connector is the one the target actually replies to and who genuinely rates you — not the most senior name.

Step 5 — Hand off

  • If a warm path exists → pass the top path + the named introducer to the warm-intro skill to write the ask + forwardable blurb.
  • If no warm path in the founder's own network (common for first-time / cross-border founders) → don't jump straight to cold. Widen the surface: accelerators, scouts, syndicates, operator angels, communities, give-to-get, build-in-public all manufacture warm intros — see references/expand-warm-surfaces.md, and vet any scout/FA channel with the fundraising-agents skill.
  • Only once warm surfaces are genuinely exhausted → the strongest direct route (an open-DM partner who posts about your space, a firm's submit page) → hand to cold-email / cold-call, anchored on the specific person and thesis.
  • Always note the contact route + DM-open/submit status you found, and a reminder to re-verify it right before reaching out.

Output format

  • The person: name, title, why them (sector/stage fit), public footprint, source links, confidence flag.
  • Ranked warm paths: introducer → relationship → trust×reachability score → how the founder knows them (Tier 1/2 source).
  • Recommended next move: the single best path, and which skill to use next.
  • Coverage note: which data the founder did/didn't provide (e.g., "no LinkedIn export yet — here's how to get it and what it would unlock").

Bilingual / China market (中文)

If the founder is raising in the China market, read shared/references/china-market-playbook.md and use references/zh-talk-tracks.md. LinkedIn is effectively gone in China and WeChat has no clean contact export, so do not ask a China founder for a LinkedIn CSV. Instead have them fill the focused contact sheet (references/contact-sheet-template.csv) from their phone contacts export, CamCard/名片全能王 export, alumni rosters, and WeChat tags — then mine that. Full how-to (中文) is in references/exporting-your-data.md. Online paths: 脉脉 (Maimai), 微信好友/群, 校友圈; X applies only for 欧美 investors.

Refuse / push back when

  • Asked to scrape LinkedIn or read the connection graph via API — explain the authorized export route instead.
  • Asked to guess or obtain personal emails to mass-contact — refuse; route to a warm path or a public submit form.
  • Asked to assert a "mutual connection" that isn't backed by the founder's data or a real source — never fabricate a relationship.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.