5 of 5 free runs left
๐ง People and company research
Sourced evaluation of real people & companies
These methods check real people and companies against sourced records. Below are the real request bodies and, where we have run them, trimmed real responses (marked as such); where a response is not a captured run it says so. No key needed to read them; get a free key to run live calls. More endpoints are further down; every one is included on every plan.
โถ Try it live โ free, no key. 5 runs per visitor a day, the same count as everywhere on this site.
score_intentโถ live ยท free
POST /v1/score_intent
Score the real buying intent of a message. Edit the text, then run it live.
Input
compare_subjectsโถ live ยท free
POST /v1/compare_subjects
Compare two real people side by side on sourced evidence. Two well-known names pre-filled.
Subjects
find_contactโถ live ยท free
POST /v1/find_contact
Resolve a named person to a verified identity and reachable contact. Run it live.
Name
Company (optional)
compare_subjects5 credits / subject
POST /v1/compare_subjects
Compare 2โ10 real people side by side on sourced evidence, context-aware. This is a real response, trimmed ("..." marks cut text).
Sample request
{
"subjects": [
{ "name": "Paul Graham", "company": "Y Combinator" },
{ "name": "Marc Andreessen", "company": "Andreessen Horowitz" }
],
"context": "general"
}
Real response (trimmed)
{
"success": true,
"version": 3,
"people": [
{ "name": "Paul Graham", "company": "Y Combinator", "photo_url": "", "summary": "",
"foxscore": 78,
"foxscore_breakdown": { "achievement": 84, "expertise": 77, "trajectory": 80,
"credibility": 70, "fit_context": 73 } },
{ "name": "Marc Andreessen", "company": "Andreessen Horowitz", "photo_url": "",
"summary": "Marc Andreessen is a software engineer, startup founder (Netscape, Opsware, Ning), and co-founder of the venture capital firm Andreessen Horowitz.",
"foxscore": 76,
"foxscore_breakdown": { "achievement": 80, "expertise": 75, "trajectory": 83,
"credibility": 64, "fit_context": 73 } }
],
"dimensions": [
{ "id": "knowledge_depth", "name": "Knowledge Depth", "category": "expertise",
"scores": { "Paul Graham": 8, "Marc Andreessen": 8 },
"notes": { "Paul Graham": "Deep expertise in startups, programming, and writing with broad intellectual curiosity.",
"Marc Andreessen": "Deep expertise in software engineering, venture capital, and technology trends." } },
{ "id": "industry_experience", "name": "Industry Experience", "category": "expertise",
"scores": { "Paul Graham": 9, "Marc Andreessen": 9 },
"notes": { "Paul Graham": "Extensive experience in tech startups, venture capital, and entrepreneurship.",
"Marc Andreessen": "Over 30 years in tech as founder, investor, and operator across multiple companies." } }
// ... 28 more dimensions (30 in all)
],
"key_insights": [
"Both Paul Graham and Marc Andreessen possess deep expertise and experience in technology and startups, scoring high on knowledge depth and industry experience.",
// ... 4 more
],
"common_connections": { "shared_companies": [], "shared_influencers": [], "shared_schools": [], "overlap_score": 0 },
"credits_used": 10
}
candidate_compare5 credits / candidate
POST /v1/candidate_compare
Compare 2โ10 candidates side by side against a role brief, on sourced evidence, for outreach and interview prep.
Sample request
{
"candidates": [
{ "name": "Maria Alvarez", "company": "Northwind Pay" },
{ "name": "Devin Okafor", "company": "Ledgerly" },
{ "name": "Sana Iqbal", "company": "Fintech Labs" }
],
"jobDescription": "Senior Backend Engineer (payments)"
}
Response (shape; credits_used is 15 for these three)
Same fields as the compare_subjects response above (same comparison engine, the
dimensions follow this framing):
people[] name, company, foxscore, foxscore_breakdown
dimensions[] 30 of them: id, name, category, scores{ per person }, notes{ per person }
key_insights[], common_connections{}, credits_used
exec_compare5 credits / exec
POST /v1/exec_compare
Compare 2โ10 executives in a leadership / competitor-exec frame, on sourced evidence.
Sample request
{
"executives": [
{ "name": "Jordan Pierce", "company": "Quanta" },
{ "name": "Lena Voss", "company": "Mapline" }
],
"context_note": "Heads of Product at competing analytics startups"
}
Response (shape; credits_used is 10 for these two)
Same fields as the compare_subjects response above (same comparison engine, the
dimensions follow this framing):
people[] name, company, foxscore, foxscore_breakdown
dimensions[] 30 of them: id, name, category, scores{ per person }, notes{ per person }
key_insights[], common_connections{}, credits_used
vet_person30 credits ยท 0 if nothing could be checked
POST /v1/vet_person
Check a person's stated claims against the public registers that settle them. One verdict per claim, the register that decided it, the exact search run, and a list of what was not checked. Not a background check.
Sample request
{ "name": "Paul Graham", "company": "Y Combinator" }
Real response (trimmed: "..." marks cut text)
{
"ok": true,
"subject": { "name": "Paul Graham", "email": null, "title": "at Y Combinator",
"employer": "Y Combinator", "links": [] },
"verdict": "partially_verified",
"verdict_sentence": "This does not read as fully verified because ... Everything we did check is listed below with the exact search, so you can finish the rest yourself.",
"claims": [
{ "claim": "The person named is Paul Graham.", "verdict": "could_not_verify",
"decided_by": "no register",
"ceiling_reason": "no public register settles an identity claim, so this cannot go higher than corroborated however many sources agree." },
{ "claim": "Paul Graham holds a role at Y Combinator.", "verdict": "partially_verified",
"decided_by": "the employer's own website", "ceiling_reason": "..." }
],
"findings": [
{ "register": "the employer's own website",
"search_performed": "Searched ycombinator.com for pages naming \"Paul Graham\".",
"outcome": "confirmed", "confidence": "inferred",
"record": "Recent Posts By Paul Graham: ...",
"record_url": "https://www.ycombinator.com/blog/author/paul-graham",
"match_strength": "identified", "supports_claim": "partial" }
// ... 1 more
],
"not_checked": [
{ "source": "SEC EDGAR company filings", "why": "No SEC-registered company matching \"Y Combinator\" was found, ...",
"check_it_yourself": "https://efts.sec.gov/LATEST/search-index?q=" }
// ... 3 more: photograph origin, identity documents, direct confirmation from the employer
],
"counts": { "registers_searched": 2, "claims_scored": 2, "not_checked": 4 },
"ceiling": "No public register holds identity itself, so nothing here concludes that a person is who they say they are ...",
"credits_used": 30
}
candidate_evaluate50 credits ยท 0 if unfindable
POST /v1/candidate_evaluate
Deep multi-source evaluation of one candidate against a role. Always asynchronous: you get a job id (HTTP 202) and poll for the result.
Sample request
{
"candidateName": "Priya Nadar",
"jobDescription": "Staff Data Scientist",
"company": "Example Analytics"
}
Response, HTTP 202 (the real envelope)
{ "job_id": "<uuid>", "poll_url": "/v1/jobs/<uuid>", "status": "pending" }
Then GET /v1/jobs/{job_id}
{ "status": "pending | running | done | error",
"result": { ... }, // when done
"credits_used": 50, // 0 when the person could not be found
"error": "..." } // when error
result carries: success, evaluation_id, candidate, job, overall_fit_score (1-100),
hiring_recommendation, technical_competency, professional_profile,
personality_behavioral, risk_assessment, gap_analysis, foxscore.
If too little public information exists: insufficient_data: true and no charge.
compare_people5 credits / subject
POST /v1/compare_people
Rank 2+ people side by side on 30 dimensions, context-aware.
Sample request
{
"subjects": [
{ "name": "Theo Marsh", "company": "Marsh Studio" },
{ "name": "Camila Reyes", "company": "Reyes Creative" }
],
"context": "general"
}
Response (shape; credits_used is 10 for these two)
Same fields as the compare_subjects response above (same comparison engine, the
dimensions follow this framing):
people[] name, company, foxscore, foxscore_breakdown
dimensions[] 30 of them: id, name, category, scores{ per person }, notes{ per person }
key_insights[], common_connections{}, credits_used
get_investor_report30 credits ยท 0 on thin data
POST /v1/get_investor_report
Snapshot report on an investor: track record, portfolio, co-investors, public thesis, red flags. Free when only limited public data is found.
Sample request
{ "name": "Jane Example", "firm": "Example Ventures" }
Response (shape from the report schema, not a captured run)
{
"dossier": {
"name", "title", "firm", "stage_focus": [...], "avg_check_size",
"contact": { "email", "twitter", "linkedin", "website" },
"investment_history": { "companies": [{ "name", "round", "amount", "year", "outcome" }], "summary" },
"portfolio_analysis", "coinvestor_patterns", "deal_structure", "board_activity",
"public_thesis": { "statements": [{ "quote", "source" }], "excited_sectors", "founder_criteria" },
"founder_sentiment", "red_flags": { "items": [{ "description", "severity", "source" }], "overall_severity" },
"personality", "data_quality_score": 1-10
},
"data_quality_score": 1-10,
"credits_used": 30
}
Want to run these live? Get a free key ยท or connect over MCP from the Augment page. Browse the full endpoint catalog below โ no key needed.