agentleFS
Overdrive-ConsultingOverdrive-Consulting/vc-fundraising-skill/.claude/skills/find-vcs/SKILL.md

find-vcs

Skill4 starsChanged 8 months agoMITView on GitHub

Find venture capital firms that match a startup's fundraising brief. Searches VC databases and firm websites using WebSearch and Apify scraping. Triggers on: "find vcs", "find investors", "search for funds", "which vcs invest in", "step 2", "next step", "continue to find vcs".

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Prefer Exa over Apify for portfolio and firm research — it handles rendering and returns cleaner structured content.

    line 25
  2. Very specific:

    Read data/normalized/company_profile.json.

    line 29
  3. Distinctive:

    Read references/vc-databases.md before searching.

    line 35
  4. Very specific:

    Use web_search_exa to find candidate firms.

    line 39
  5. Distinctive:

    Fallback to Apify: if Exa returns incomplete portfolio data for a specific firm:

    line 81
  6. Distinctive:

    Check size (e.g. $50M+ checks for a $500K raise)

    line 91
  7. Very specific:

    Write an array to data/raw/fund_candidates.json conforming to schemas/fund_candidates.schema.json.

    line 111
  8. Very specific:

    After writing data/raw/fund_candidates.json, present a summary table to the founder:

    line 119
  9. Very specific:

    Do not proceed to map-investors until the founder confirms or adjusts the list.

    line 142

≈ “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

Find VCs

Pipeline: clarify-thesis → find-vcs → map-investors → enrich-vc-contacts → score-vc-fit → draft-investor-outreach

You are step 2 of 6. Your job is to identify 15–30 candidate VC firms that plausibly match the startup's fundraising profile.

Tools

  • Exa MCP (web_search_exa) — primary VC database discovery
  • Exa MCP (company_research_exa) — deep research on a specific firm: portfolio, thesis, recent activity
  • Exa MCP (crawling_exa) — fetch and extract content from firm/portfolio pages
  • Apify MCP (apify/cheerio-scraper) — fallback for structured portfolio page extraction (JS-light pages)
  • Apify MCP (apify/website-content-crawler) — fallback for JS-heavy firm websites
  • WebSearch / WebFetch — fallback for anything Exa doesn't cover

Prefer Exa over Apify for portfolio and firm research — it handles rendering and returns cleaner structured content. Use Apify only when Exa returns incomplete results for a specific page.

Input

Read data/normalized/company_profile.json. If the file does not exist or required fields are null, stop and ask the founder to complete clarify-thesis first.

Search strategy

Step 0: Read the VC databases reference

Read references/vc-databases.md before searching. It lists recommended sources, search patterns, and Apify actors for each database.

Step 1: Discover VC firms using Exa

Use web_search_exa to find candidate firms. Run at least 5 queries from different angles. Tailor to the startup's sector, stage, and geography.

SourceExa query
OpenVCsite:openvc.app [sector] [stage] investors
Crunchbasesite:crunchbase.com/organization [sector] VC fund [stage]
Signal by NFXsite:signal.nfx.com [sector] [stage]
Notion VC lists"vc list" [sector] [stage] site:notion.so
Airtable VC lists"vc list" [sector] [stage] site:airtable.com
General[sector] seed fund OR venture capital investors [geography]
Portfolio adjacency"[comparable_company]" investors OR "backed by" OR "portfolio"

Deduplicate by fund name across all queries.

Step 2: Research each firm's portfolio using Exa

For each candidate firm found in Step 1, use company_research_exa to pull structured information:

company_research_exa("[fund_name] venture capital")

This returns: investment thesis, sector focus, stage, portfolio companies, check size signals, and recent activity — without needing to scrape individual pages.

Then use crawling_exa to fetch the firm's portfolio page directly for the full portfolio list:

crawling_exa("[fund_website]/portfolio")

or

crawling_exa("[fund_website]/companies")

Extract:

  • Portfolio company names (to check for comparable_companies matches)
  • Sector signals from portfolio descriptions
  • Stage signals from portfolio company maturities
  • Any stated check size or investment criteria

Fallback to Apify if Exa returns incomplete portfolio data for a specific firm:

  • apify/cheerio-scraper — for structured portfolio grids on JS-light pages
  • apify/website-content-crawler — for JS-heavy firm websites

Step 3: Filter candidates

Exclude firms that are clearly out on:

  • Stage (e.g. growth-stage only fund for a pre-seed company)
  • Geography (e.g. LATAM-only fund for a US company)
  • Sector (e.g. biotech-only fund for a SaaS company)
  • Check size (e.g. $50M+ checks for a $500K raise)

Flag partial fits with a note in possible_concerns.

What counts as a match

Evaluate each firm on:

  • sector fit
  • stage fit
  • geography fit
  • check size fit
  • lead/follow fit
  • portfolio adjacency (invested in comparable_companies or direct competitors)
  • recent investment activity (last 24 months)
  • thematic alignment with thesis_keywords
  • explicit investor preferences from the profile

Output

Write an array to data/raw/fund_candidates.json conforming to schemas/fund_candidates.schema.json.

Target 15–30 firms. Fewer is fine if quality is high. Do not pad with weak fits.

Assign each fund a fund_id as a lowercase hyphen-separated slug of the fund name (e.g. sequoia-capital). This is the canonical join key used in all later steps.

Review checkpoint

After writing data/raw/fund_candidates.json, present a summary table to the founder:

Found [N] candidate VC firms for [company_name]:

| # | Fund | Stage | Sector focus | Check size | Confidence |
|---|------|-------|-------------|-----------|-----------|
| 1 | ... | ... | ... | ... | high/medium/low |
...

[If any partial fits]:
  ⚠ Partial fits flagged: [list with concerns]

[If evidence was thin for some firms]:
  Note: Evidence confidence is low for: [list] — worth verifying manually.

Does this list look right? You can:
  - Confirm to proceed to step 3 (mapping people at each firm)
  - Remove specific firms: "remove [fund name]"
  - Add a firm manually: "add [fund name] at [website]"
  - Narrow the list: "only keep firms with high confidence"

Do not proceed to map-investors until the founder confirms or adjusts the list.

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.