enrich-vc-contacts
Skill4 starsChanged 8 months agoMITView on GitHub
Enrich public professional contact details for each mapped investor. Uses icypeas_official/bulk-email-finder for batch email lookup, and Exa for any missing LinkedIn URLs. Clearly labels all inferred data. Triggers on: "enrich contacts", "find emails", "get contact details", "step 4", "next step", "continue to enrich contacts".
In this repository8 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- clarify-thesisSkill
- draft-investor-outreachSkill
- export-csvSkill
- find-vcsSkill
- map-investorsSkill
- resetSkill
- score-vc-fitSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 0 are distinctive, and 7 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Read data/raw/investor_list.json.
- Very specific: line 44
Run Apify actor icypeas_official/bulk-email-finder with the prospect CSV.
- Very specific: line 65
Use Exa MCP people_search_exa:
- Very specific: line 97
Write an array to data/normalized/investor_contacts.json conforming to schemas/investor_contact.schema.json.
- Very specific: line 99
Every record must include contact_tier and source_type, even if email is not found.
- Very specific: line 109
After writing data/normalized/investor_contacts.json, present a summary to the founder:
- Very specific: line 137
Do not proceed to score-vc-fit until the founder confirms or adjusts the contacts.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.
Files
Enrich VC Contacts
Pipeline: clarify-thesis → find-vcs → map-investors → enrich-vc-contacts → score-vc-fit → draft-investor-outreach
You are step 4 of 6. Your job is to enrich contact details for each person in investor_list.json — primarily emails, with LinkedIn URLs as fallback.
Tools
- Apify MCP (
icypeas_official/bulk-email-finder) — primary: batch email lookup by name + company domain - Exa MCP (
people_search_exa) — fill in any missing LinkedIn URLs - WebSearch / WebFetch — fallback for checking firm bio pages
Input
Read data/raw/investor_list.json. If the file does not exist, stop and ask the founder to complete map-investors first.
Step 1: Prepare the prospect list
From investor_list.json, build a list of { name, domain } pairs — one per investor.
Extract the domain from fund_website in fund_candidates.json (e.g. sequoiacap.com from https://www.sequoiacap.com). Strip www. and the protocol.
The actor accepts up to 5,000 rows. Format as CSV:
firstName,lastName,domain
Sarah,Smith,sequoiacap.com
James,Lee,benchmark.comSplit first and last name from the name field. If only a single name token is available, put it in firstName and leave lastName empty.
Step 2: Run bulk email finder
Run Apify actor icypeas_official/bulk-email-finder with the prospect CSV.
The actor returns:
email— found email addressconfidence— confidence score (0.0–1.0)status—found,not_found, orrisky
Map results back to investors by firstName + lastName + domain.
Assign contact_tier based on the result:
| Actor status | confidence | contact_tier | email_confidence |
found | ≥ 0.8 | confirmed | actor value |
found | < 0.8 | inferred | actor value |
not_found or risky | — | not_found | null |
Step 3: Fill missing LinkedIn URLs
For any investor still missing a linkedin URL after map-investors:
Use Exa MCP people_search_exa:
"[name]" "[fund_name]" venture capitalExtract the LinkedIn URL from the result. If not found, leave linkedin: null — the investor can still be contacted via their firm contact page.
Step 4: Check firm bio pages (spot check only)
For investors where both email and LinkedIn are still missing after Steps 2–3:
WebSearch: "[name]" site:[firm_website]
Or WebFetch {firm_website}/team and scan for the person's name. Some firms list emails directly on their team page — if found, set contact_tier: "confirmed".
This step is a spot check, not a full scrape. Do not spend time on firms where Step 2 already returned results.
Contact information rules
Allowed:
- Email returned by
icypeas_official/bulk-email-finderwith statusfound→ tier based on confidence - Email found directly on a public firm page →
contact_tier: "confirmed",email_confidence: 1.0 - Public LinkedIn URL
- Firm bio page URL
Not allowed:
- Personal email addresses
- Non-public phone numbers
- Presenting a low-confidence email as confirmed
Output
Write an array to data/normalized/investor_contacts.json conforming to schemas/investor_contact.schema.json.
Every record must include contact_tier and source_type, even if email is not found.
Records with contact_tier: "not_found" are still included — they are contacted via LinkedIn.
Set source_type:
"public"— email found on a firm page"probabilistic"— email returned by bulk-email-finder"not_found"— no email found
Review checkpoint
After writing data/normalized/investor_contacts.json, present a summary to the founder:
Enriched [N] contacts across [M] firms for [company_name]:
Contact quality breakdown:
✓ Confirmed email (high confidence): [N] contacts
~ Inferred email (lower confidence): [N] contacts
✗ No email found — LinkedIn only: [N] contacts
| Fund | Name | Email | Confidence | LinkedIn | Tier |
|------|------|-------|-----------|---------|------|
| ... | ... | ... | 0.92 | yes/no | confirmed |
...
[If many low-confidence emails]:
Note: [N] emails have confidence below 0.8. Treat as best-guess — verify before sending or use LinkedIn as primary channel.
[If many not_found]:
Note: [N] contacts have no email. LinkedIn DM is the recommended channel for these.
Does this look right? You can:
- Confirm to proceed to step 5 (scoring fit and ranking)
- Correct an email: "update email for [name] to [email]"
- Remove a contact: "remove [name]"
- Skip enrichment for a firm: "skip [fund name]"Do not proceed to score-vc-fit until the founder confirms or adjusts the contacts.
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
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.

