x402 / Agentic.Market
Agent Research Desk
Agent Research Desk is a pay-per-call trust and diligence API for AI agents. It helps agents make first-pass decisions before they contact, approve, buy from, rely on, or escalate a company, vendor, claim, or lead.
Agent Guide
https://agent-research-brief-api-production.up.railway.app/agents
Agent Capabilities
https://agent-research-brief-api-production.up.railway.app/agent-capabilities.json
Free Trust Preview
https://agent-research-brief-api-production.up.railway.app/trust-preview
Start Here
https://agent-research-brief-api-production.up.railway.app/start
Company Intel
https://agent-research-brief-api-production.up.railway.app/company-intel
Company Watch
https://agent-research-brief-api-production.up.railway.app/company-watch
Health
https://agent-research-brief-api-production.up.railway.app/health
Status
https://agent-research-brief-api-production.up.railway.app/status
Docs JSON
https://agent-research-brief-api-production.up.railway.app/docs.json
Tester Guide
https://agent-research-brief-api-production.up.railway.app/tester
Activation Path
https://agent-research-brief-api-production.up.railway.app/activation
Enterprise
https://agent-research-brief-api-production.up.railway.app/enterprise
Dashboard
https://agent-research-brief-api-production.up.railway.app/dashboard
Agent-First
One public entry path for other agents
- Start at /agents so the buyer agent sees one entry path instead of the whole product surface.
- Inspect /agent-capabilities.json or /docs.json first, then use /trust-preview or /start before spending.
- Buy /api/company-intel when the named company deserves a real trust decision now.
- Turn that same subject into /api/company-watch when repeat value depends on meaningful change over time.
Controlled Activation
Fastest path for the first real buyer wave
- Start with the free Trust Preview for one named company so the first step feels useful before money moves.
- Move into Company Intel only when the preview says the target is real enough to justify a trust decision now.
- Upgrade to Company Watch when the same company is likely to matter again and the next decision depends on meaningful change over time.
- Only widen into Source Check, Vendor Risk, or Deep Diligence when the blocker becomes specific enough to justify the extra spend.
Best For
Who buys this service
- Procurement, trust, or operator agents screening vendors before a pilot or approval
- Research agents that need deterministic JSON instead of open-ended narrative output
- Growth or outreach agents that want to qualify a target account before spending human time
Why It Converts
What makes the product ladder useful
- Agents can start cheap, then escalate only when the evidence gap is real.
- Each route has a specific trust or diligence job instead of vague research sprawl.
- Premium deep diligence is framed as decision prep, not a final verified verdict.
Trust Standard
First-Pass Trust Standard
- Every paid response now includes a trustProfile block that says what the result is safe to use for and what it is not.
- Confidence shows how strong the first-pass read is. trustProfile shows where the safe-use boundary is.
- Provided-source reads are called out clearly, but they are not presented as independent verification.
- High-stakes decisions should still verify the strongest claim, counterparty, or control artifact before acting.
Premium Route
Deep Diligence Plan
This is the highest-value route in the service at $0.50. It is meant for meaningful trust decisions, not casual exploration.
- The subject is tied to procurement, partnership, treasury exposure, or another decision with real downside.
- A quick screen is no longer enough because the buyer needs to know what proof is still missing.
- The service should produce a verification map that another agent or human can act on immediately.
- This is decision prep, not a final legal, financial, compliance, or investment opinion.
Enterprise
Enterprise Trust Desk
Enterprise Trust Desk is the private, higher-touch version of Agent Research Desk for teams that want custom diligence workflows, higher-volume buying, and operator support without changing the working public x402 product.
Path: /enterprise
Buying Guide
Choose Your Starting Point
These are the fastest ways for an agent to decide what to buy first instead of guessing.
check_opportunity
/api/opportunity-check
Buy this when: You are unsure which product to buy or whether the topic is even worth paying for yet.
Why: Start with the cheapest screen, then escalate only if the signal is real.
Price: $0.01
create_company_intel
/api/company-intel
Buy this when: You already know the company or account and need a fast trust-and-growth read.
Why: This is the fastest path to a usable snapshot before outreach or deeper diligence.
Price: $0.05
create_deep_diligence_report
/api/deep-diligence
Buy this when: The decision is meaningful enough that proof gaps could change the outcome.
Why: Use the premium plan when the buyer needs a serious verification map, not just a quick screen.
Price: $0.50
check_source_claim
/api/source-check
Buy this when: One specific claim, source, or proof point is the blocker.
Why: Do the cheapest targeted verification first instead of buying a larger report too early.
Price: $0.05
Free Trust Preview
Get the pre-buy verdict before you spend on Company Intel
This is the new front door for one named company. It tells the buyer whether the paid company snapshot is justified now or whether they should stay cheaper first.
preview_company_trust
/api/trust-preview before /api/company-intel
Trust Preview is the free front door for one named company. It tells the buyer whether a paid Company Intel call is justified now or whether they should stay cheaper first.
- You already have one named company but are not ready to pay without a better reason.
- You want a fast pre-buy verdict before committing to Company Intel.
- You want the product to explain what the full paid snapshot would unlock before you spend.
What it returns
Why it changes the first-buy step
- A free verdict on whether Company Intel looks worth buying now
- A clear buy-now handoff into the paid route
- A safer cheaper fallback when the company still looks too thin
- The likely second paid move if the paid company read lands
Buy-now handoff: /api/company-intel at $0.05.
Stay cheaper when needed
Do not force the full paid route too early
- You want the full company decision already. This is only the pre-buy gate.
- You need final diligence, approval, or verified outside facts.
- You do not have one named company yet.
Fallback: /api/opportunity-check at $0.01.
First Buy Route Picker
Pick the safest first paid call
This is the fastest way to avoid overbuying. If you want the service to tell you the best first route directly, start at /start or call /api/route-advisor.
check_opportunity
/api/opportunity-check
Buy this when: I only want the cheapest first paid test.
Why: Use the $0.01 screen when you are still deciding whether the target deserves more spend or you simply want to prove the paid flow with low risk.
Price: $0.01
Boundary: Do not skip this step unless you already know the exact blocker.
create_company_intel
/api/company-intel
Buy this when: I already have one named company and want a fast trust snapshot.
Why: Use Company Intel when the question is already about one real company and you want posture, signals, and the next trust question fast.
Price: $0.05
Boundary: If you are still unsure the target deserves more than a cheap screen, start with Opportunity Check first.
assess_vendor_risk
/api/vendor-risk
Buy this when: I already know the vendor and need an approval or rollout read.
Why: Use Vendor Risk when the decision is about pilot, procurement, partnership, or vendor approval rather than general research.
Price: $0.05
Boundary: Do not use this as your first purchase if you only want to test the paid flow cheaply.
check_source_claim
/api/source-check
Buy this when: One claim or source is the blocker.
Why: Use Source Check when a single statement or source is what stands between you and the next decision.
Price: $0.05
Boundary: Do not buy a broader route if one focused proof question is the real issue.
create_deep_diligence_report
/api/deep-diligence
Buy this when: Missing evidence could change a real decision.
Why: Use Deep Diligence only when you need a serious proof-gap map and the next move depends on missing evidence.
Price: $0.50
Boundary: Do not use this as a first purchase unless the decision is already meaningful enough to justify the premium step.
Traffic Leader
Why most buyers are landing on Company Intel
This is the busiest paid route because buyers usually already have one named company. The free Trust Preview now exists to make that first paid step feel safer and more justified.
create_company_intel
/api/company-intel at $0.05
Company Intel is the strongest first paid route when the buyer already has one named company and needs a fast trust snapshot before outreach, approval, or deeper diligence.
- You already know the company name and want a real first-pass trust read, not a broad memo.
- You need posture, signals, likely proof gaps, and the cleanest next route in one paid call.
- You want to know whether this company deserves vendor-risk, source-check, or deep-diligence next.
What the buyer gets
Why the route is worth buying
- A company posture read and likely buyer motion
- Trust signals, watch items, and obvious proof gaps
- Clear next actions and follow-up questions
- The best next paid route if the company is worth deeper work
Do not start here
When to stay cheaper or go deeper
- You are still unsure the target deserves more than the $0.01 screen.
- One specific claim is the blocker and a focused source check would answer it faster.
- The decision is already serious enough that only a premium proof-gap map will do.
If you are unsure: start at /start.
Recurring Trust Lane
Keep the same company alive instead of starting over
Company Watch is the compounding layer. Use it after the first company read when the real value is spotting meaningful changes and turning them into the next justified paid step.
create_company_watch
/api/company-watch at $0.20
Company Watch is the recurring trust lane for one named company. It keeps the target warm between outreach, approval, partnership, or diligence decisions instead of making the buyer restart from zero later.
- You already ran a first company read and now want the desk to keep the same company alive between real decisions.
- The company matters enough that you want change-sensitive trust monitoring, not another one-shot report.
- You want the next spend to be triggered by proof changes, not by guesswork or memory.
What the watch does
Why this changes the business loop
- A live watch baseline for one named company
- A refresh cadence and alert sensitivity matched to the watch goal
- Proof gaps, trigger conditions, and the next justified follow-on route
- A cleaner repeat loop around the same trust question instead of a reset
Follow-on lane: Use Company Watch after a first company read when the best next move depends on meaningful changes over time, not another fresh reset.
Do not start here
Use this after the first company decision exists
- You do not know yet whether the company deserves more than a free preview or a cheap first paid screen.
- You only need a one-time company snapshot right now and do not need a recurring watch queue.
- One claim is already the only blocker and a focused source check would answer it faster.
Best handoff: start with /company-intel or the free preview at /trust-preview.
Why Buyers Trust The Output
Every paid response now carries its own trust boundary
Buyers do not have to infer what the result means. The response shape now says what it is safe to use for, what still needs proof, and whether a provided source was read directly.
Always Included
Core trust fields
- confidence: how strong the first-pass read is
- trustProfile: safe-use boundary, proof burden, and verification status
- decisionSignals: urgency, proof-gap severity, and the safest next operating move
- sourceAttribution: where the next proof should come from
Safe Use
What buyers can rely on
- A conservative first-pass judgement instead of a vague memo
- Clear next actions and proof gaps before a higher-stakes decision
- A cleaner handoff into a deeper route when the evidence gap is real
Not Claimed
What the desk still refuses to overpromise
- No final legal, compliance, investment, or audit verdicts
- No pretending recommended sources were already verified
- No hiding uncertainty behind polished language or generic confidence scores
Repeat Use
Every paid response now points to the next paid move
The desk should not feel like a one-off report. It should tell the buyer what to buy next while the same subject is still warm.
Second Paid Call Standard
How the ladder stays moving
- Every paid response now includes commercial.repeatUsePlan so the buyer can see the single best next route immediately.
- The best next paid step should stay on the same subject while the trust question is still warm.
- If the next blocker is still unclear, the safest fallback is the cheap front door instead of overbuying the next route.
Inside The Response
What the buyer gets back
- commercial.recommendedOffers for the strongest follow-on routes
- commercial.repeatUsePlan for the single best next paid move
- timing guidance so the buyer knows whether to continue now or after one outside step
Repeat-Buyer Logic
How to avoid dead-end reports
- Stay on the same subject while the trust question is still live.
- Escalate only when the next blocker is clearer than the last one.
- If the next blocker is still fuzzy, go back to the cheap front door instead of overbuying.
Live Status
Deployment Snapshot
Use /status for the machine-readable mode, network, and tool map.
Modepaid
Configured Networkeip155:8453
Facilitator
https://api.cdp.coinbase.com/platform/v2/x402MCP Command
npm run mcpDiscovery MetadataBazaar metadata declared for every paid endpoint
By Invite Source
Source-aware first buys
- Warm technical tester: start with /api/opportunity-check at $0.01, then escalate to /api/company-intel only when the cheap screen says the target deserves more spend and the next question is about one real company
- x402 or Base builder: start with /api/opportunity-check at $0.01, then escalate to /api/vendor-risk only when the buyer already knows the vendor and the real blocker is trust, rollout risk, or procurement readiness
- MCP or agent-tool builder: start with /api/company-intel at $0.05, then escalate to /api/deep-diligence only when the first-pass company read still leaves proof gaps that could change a real decision or approval flow
- Base ecosystem builder: start with /api/opportunity-check at $0.01, then escalate to /api/company-intel only when the cheap screen suggests the company or operator is worth a proper trust-and-posture read
- Manual direct outreach: start with /api/company-intel at $0.05, then escalate to /api/vendor-risk only when the buyer already knows the company is interesting and now needs a sharper trust or control read
Paywall Guidance
What a clean unpaid challenge should do
- Show the listed route price instead of leaving the buyer guessing.
- Restate what the route is actually for before the buyer retries.
- Point unsure buyers to /start before they overbuy.
- Point unsure buyers back to /api/opportunity-check at $0.01.
- Give the busiest route a clearer explanation at /company-intel.
- Point buyers to /docs.json when they need the schema before retrying.
Operator Control Room
Single private admin home
A single private operator home combines the live desk pulse, Bazaar scout digest, outreach funnel, and the next move so you do not need multiple admin tabs open.
- Fast operator summary for the live service
- Change-only private inbox from the market loop
- Aggregate tester and outreach signal in one place
Operator Briefs
Short private memos
Private daily and weekly operator briefs turn the live desk, Bazaar scout, and tester funnel into a short memo so operators can decide what to fix next without digging through raw telemetry.
- Short daily and weekly operator summaries
- What is working, what is risky, and what to do next
- Private metrics only, with no public buyer-level exposure
Market Scout
Internal Bazaar loop
A private Bazaar / Agentic Market scout watches discovery, merchant visibility, and adjacent route density, then feeds the results into the operator control room.
- Merchant visibility for the seller wallet
- Route-by-route overlap with adjacent Bazaar services
- Crowded lanes, whitespace, and suggested positioning moves
Private Operator Routes
Where the scout reports
- /admin/control-room for the single private control room
- /api/admin/control-room for the combined admin JSON snapshot
- /admin/briefs for private daily and weekly operator briefs
- /api/admin/operator-brief for private brief JSON output
- /admin/market-scout for the private market view
- /api/admin/market-scout for the private market JSON snapshot
- /admin/dashboard for the full private request stream
Launch Proof
Verified Production Payment
This is the paid Base mainnet proof used to confirm the live launch candidate.
Tested Endpoint
POST /api/company-intel
Seller Wallet
0x83cE8ec409Ef94F7d909ecDCd8cC578741b0231D
Network
eip155:8453
Facilitator
https://api.cdp.coinbase.com/platform/v2/x402
Paid Test Date
2026-04-20
Historical Proof Tag
v1.0-mainnet-paid-discovery-confirmed
Current Release Tag
v1.8.0-agent-acquisition-loop
Discovery
Visible
What Passed
Verified checks
- Unpaid request returned HTTP 402.
- Paid request returned HTTP 200 with structured JSON.
- USDC settled to the seller wallet on Base mainnet.
- Merchant and Bazaar discovery became visible after the paid request.
Agent Workflows
Common Buying Paths
These are the product ladders that should be easiest to understand and easiest to buy from programmatically.
Vendor approval flow
opportunity-check → vendor-risk → deep-diligence
Move from uncertainty to a trust decision without buying the expensive report too early.
/api/opportunity-check· $0.01/api/vendor-risk· $0.05/api/deep-diligence· $0.50
Account qualification flow
company-intel → company-watch → lead-brief
Turn a named company into a clearer outreach or diligence path.
/api/company-intel· $0.05/api/company-watch· $0.20/api/lead-brief· $0.05
Claim-to-proof flow
source-check → company-intel → deep-diligence
Test a shaky statement first, then widen the work only if it still matters.
/api/source-check· $0.05/api/company-intel· $0.05/api/deep-diligence· $0.50
Recurring trust watch flow
company-intel → company-watch → vendor-risk
Keep one named company under a live trust queue so the next decision does not restart from zero.
/api/company-intel· $0.05/api/company-watch· $0.20/api/vendor-risk· $0.05
For Agents
Exact Agent Flow
- Discover the service in Bazaar or call the public URL directly.
- Use /trust-preview or /start first when you want a free first pass before paying.
- Inspect /docs.json to read schemas, pricing, tool names, and example payloads.
- Send an unpaid POST to a protected route and read the HTTP 402 payment requirements.
- Retry the same request with x402 on Base mainnet and pay the listed route price starting at $0.01 USDC.
- Receive structured JSON with confidence, trust boundary, source attribution, and a clear repeat-use path for the next paid call.
Structured Output
What buyers get back
- summary
- confidence
- trustProfile
- decisionSignals
- risks
- opportunities
- nextActions
- questions
- sourceAttribution
- commercial
Supported Networks
Where payment works
- Base Sepolia for testing via eip155:84532
- Base mainnet for production via eip155:8453
- Configured facilitator: https://api.cdp.coinbase.com/platform/v2/x402
MCP Tools
Tool availability
- preview_company_trust → /api/trust-preview
- choose_best_route → /api/route-advisor
- check_opportunity → /api/opportunity-check
- create_research_brief → /api/research-brief
- create_company_intel → /api/company-intel
- create_company_watch → /api/company-watch
- check_source_claim → /api/source-check
- assess_vendor_risk → /api/vendor-risk
- create_deep_diligence_report → /api/deep-diligence
- create_lead_brief → /api/lead-brief
Privacy
Public data boundary
Agent Research Desk does not publicly expose buyer prompts, wallet-level request history, or paid response contents. Public metrics are aggregate only.
Examples
Realistic Example Payloads
These are synthetic public examples for common agent buying flows. They do not expose buyer-level history or paid response contents from real customers.
check_opportunity
Test whether an idea or target deserves deeper spend
$0.01
Use this when an agent needs a cheap yes-or-no screen before committing to a full brief, company read, or risk review.
Outcome: The response returns a pass/watch/skip verdict, the likely buyer, the trust question, the fastest proof path, and the best next paid product.
POST /api/opportunity-check
Example request
{
"subject": "Agent trust scoring for crypto vendors",
"buyer": "Crypto fund operator",
"marketType": "vendor",
"geography": "United States",
"goal": "Decide whether this topic is strong enough to package into a deeper diligence offer.",
"knownSignals": [
"Operators keep asking for a faster first-pass trust screen before pilots",
"Funds want a lightweight way to compare agent vendors before deeper diligence"
],
"constraints": [
"Keep the first offer lightweight",
"Do not assume the buyer already trusts agent vendors"
]
}Example response
{
"requestId": "c6d33af8-5a3b-4018-9b56-30f8b9dd2c91",
"createdAt": "2026-04-20T18:11:00.000Z",
"endpoint": "opportunity-check",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.01",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"subject": "Agent trust scoring for crypto vendors",
"buyer": "Crypto fund operator",
"marketType": "vendor",
"geography": "United States",
"goal": "Decide whether this topic is strong enough to package into a deeper diligence offer.",
"knownSignals": [
"Operators keep asking for a faster first-pass trust screen before pilots",
"Funds want a lightweight way to compare agent vendors before deeper diligence"
],
"constraints": [
"Keep the first offer lightweight",
"Do not assume the buyer already trusts agent vendors"
]
},
"summary": "This looks like a promising entry-point offer because the buyer pain is concrete, the trust problem is obvious, and the topic can escalate cleanly into a deeper diligence product if the first signal holds up.",
"confidence": {
"score": 0.74,
"label": "medium",
"rationale": [
"The request names a buyer, market type, geography, and a concrete packaging goal.",
"Known signals point to repeat demand instead of a purely speculative idea."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "request-only",
"evidenceStrength": "developing",
"decisionReadiness": "usable",
"proofBurden": "moderate",
"useFor": [
"Choosing whether the topic deserves a deeper paid route",
"Starting with a cheap screen before a company, vendor, claim, or lead-specific buy"
],
"notFor": [
"Treating this as final diligence or final approval",
"Claiming the target was independently verified from outside evidence"
],
"whyThisIsUseful": [
"It makes the next paid step explicit instead of forcing a blind higher-price purchase.",
"It stays conservative when proof is still thin, which makes it safer as a routing decision than as a verdict."
]
},
"decisionSignals": {
"urgency": "watch",
"spendReadiness": "screen-only",
"proofGapSeverity": "moderate",
"actNowBecause": "This is the cheapest useful way to prove whether the topic deserves deeper spend at all.",
"holdIf": "Hold the next step if the topic still feels vague or the trust question is not yet concrete.",
"escalationTrigger": "Escalate only when one company, vendor, claim, or lead becomes the obvious next target.",
"operatorCue": "Keep the first spend low until the target proves it deserves a fuller route."
},
"risks": [
"The topic may sound urgent, but buyers might still want proof that the trust problem is expensive enough to pay for.",
"A lightweight front-door offer can attract curiosity clicks if the next paid step is not clear and specific."
],
"opportunities": [
"This is cheap enough to convert uncertain buyers without forcing them into a full brief immediately.",
"The topic routes naturally into a higher-value vendor diligence workflow if the first screen lands."
],
"nextActions": [
"Test whether buyers want a fast score, a short rationale, or a direct escalation recommendation first.",
"Ask one real buyer what proof would make them pay for the next, deeper step.",
"Track which market types convert from this cheap screen into bigger products."
],
"questions": [
"What would make a buyer pay for a deeper follow-up right after this screen?",
"Is the strongest demand tied to one named vendor, one company, or a broader market category?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "Known request signals",
"url": null,
"relevance": "Initial buyer signal",
"note": "Use the signals supplied in the request as the first evidence of urgency and packaging fit."
},
{
"kind": "recommended",
"label": "Follow-up buyer interviews or repeated paid requests",
"url": null,
"relevance": "Conversion validation",
"note": "Use repeat purchase behavior or direct buyer feedback to validate whether the cheap screen deserves to stay in the ladder."
}
],
"commercial": {
"stage": "entry",
"buyerRelationship": "new",
"notes": [
"This is the lowest-friction entry product, so the main job is deciding whether the topic deserves deeper paid work.",
"The next offer should point directly to the first serious diligence or research product, not another vague idea screen."
],
"recommendedOffers": [
{
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "The topic is already framed as a vendor trust problem, so the clean next move is a structured risk review.",
"suggestedUse": "Use this when the cheap screen says the vendor question is real enough to pressure-test properly."
},
{
"endpoint": "company-intel",
"title": "Company Intel",
"path": "/api/company-intel",
"stage": "core",
"priceUsd": "$0.05",
"reason": "If the opportunity centers on one company, the next step is understanding the operator behind the pitch.",
"suggestedUse": "Run this when you need a fuller operating and trust view of the company behind the topic."
},
{
"endpoint": "research-brief",
"title": "Research Brief",
"path": "/api/research-brief",
"stage": "core",
"priceUsd": "$0.05",
"reason": "If the topic looks bigger than one vendor, turn the screen into a fuller market brief.",
"suggestedUse": "Use this when the signal is real but still needs market framing before action."
}
],
"repeatUsePlan": {
"bestNextRoute": "vendor-risk",
"bestNextTitle": "Vendor Risk",
"bestNextPath": "/api/vendor-risk",
"bestNextPriceUsd": "$0.05",
"timing": "now",
"whyNow": "The entry screen only does its job if it points to the first serious follow-up route right away.",
"buyAgainWhen": "Buy again as soon as the cheap screen says the trust question is real enough to pressure-test properly.",
"ifUnsure": "If the cheap screen still feels inconclusive, stay with /api/opportunity-check until the target deserves more spend."
}
},
"screen": {
"verdict": "pass",
"rationale": "The request is specific enough to price cheaply, and it has a clear escalation path into vendor diligence if the first signal converts.",
"likelyBuyer": "Crypto fund operator",
"urgency": "Operators are already asking for a faster first-pass trust screen before deeper vendor review.",
"trustQuestion": "Will a buyer trust a lightweight signal enough to pay for a full vendor risk review next?",
"fastestProof": "Sell the cheap screen first, then measure whether buyers immediately escalate into a vendor-focused follow-up.",
"bestNextProduct": "vendor-risk"
}
}create_company_intel
Screen a company before outreach, diligence, or partnership review
$0.05
Use this when an agent needs a fast trust-and-growth read on a company before spending time on deeper research.
Outcome: The response frames the company posture, likely buyer motion, visible trust signals, and the follow-up checks that matter most.
POST /api/company-intel
Example request
{
"companyName": "Alpine Foods",
"website": "https://example.com",
"industry": "Food distribution",
"geography": "United States",
"researchFocus": "growth",
"knownClaims": [
"Regional specialty food distributor",
"Expanding private-label program"
],
"knownSignals": [
"Hiring for operations and sales",
"Launching new product categories"
],
"questions": [
"How concentrated is demand around a few buyers?",
"What trust signals matter most to partners?"
]
}Example response
{
"requestId": "0f8f6988-dcaa-4dd7-9aef-4308b65bf3c7",
"createdAt": "2026-04-20T18:05:00.000Z",
"endpoint": "company-intel",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.05",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"companyName": "Alpine Foods",
"website": "https://example.com",
"industry": "Food distribution",
"geography": "United States",
"researchFocus": "growth",
"knownClaims": [
"Regional specialty food distributor",
"Expanding private-label program"
],
"knownSignals": [
"Hiring for operations and sales",
"Launching new product categories"
],
"questions": [
"How concentrated is demand around a few buyers?",
"What trust signals matter most to partners?"
]
},
"summary": "Alpine Foods looks like a growth-stage food distribution business where commercial momentum depends on partner trust, category expansion discipline, and operational execution more than pure brand awareness.",
"confidence": {
"score": 0.76,
"label": "high",
"rationale": [
"The request includes a company name, website, industry, geography, and explicit growth focus.",
"Known claims and signals provide enough framing for a first-pass operating view."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "provided-source-read",
"evidenceStrength": "strong",
"decisionReadiness": "decision-ready",
"proofBurden": "moderate",
"useFor": [
"Getting a first-pass company snapshot before outreach, approval, or deeper diligence",
"Comparing public posture against the most obvious proof gaps"
],
"notFor": [
"Treating this as audited company diligence or a final financial judgement",
"Claiming the company story was independently verified beyond the supplied inputs and cited sources"
],
"whyThisIsUseful": [
"A provided source is attached, so the result is anchored to a real first-party surface before manual follow-up.",
"It separates operating posture from missing evidence so buyers can decide whether to escalate or stop."
]
},
"decisionSignals": {
"urgency": "active",
"spendReadiness": "core-route",
"proofGapSeverity": "moderate",
"actNowBecause": "The company read is strongest when it quickly turns into one concrete trust, outreach, or proof follow-up.",
"holdIf": "If you still are not sure the company deserves more spend, go back to /api/opportunity-check.",
"escalationTrigger": "Buy again when trust, one public claim, or approval risk becomes the clear next blocker for the same company.",
"operatorCue": "Use the snapshot to choose whether trust, outreach, or one proof check is the real next move."
},
"risks": [
"Expansion into new categories can stretch fulfillment quality before revenue diversification pays off.",
"A trust-led sales motion may depend on references, proof of service levels, and margin discipline."
],
"opportunities": [
"Private-label growth can create margin expansion if buyer trust is already strong.",
"Operational hiring may indicate enough demand to justify deeper diligence into execution capacity."
],
"nextActions": [
"Verify concentration risk across customers, product lines, and regions.",
"Check whether growth claims are supported by hiring patterns, launches, or partner announcements.",
"Test whether the website and public messaging match the actual buyer motion."
],
"questions": [
"How concentrated is demand around a few buyers?",
"What trust signals matter most to partners?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "Company website",
"url": "https://example.com",
"relevance": "Primary positioning source",
"note": "Use the official site to compare stated capabilities against external signals."
},
{
"kind": "recommended",
"label": "Partner, hiring, and launch signals",
"url": null,
"relevance": "Momentum validation",
"note": "Use public job postings, announcements, and partner references to validate growth posture."
}
],
"commercial": {
"stage": "core",
"buyerRelationship": "new",
"notes": [
"Synthetic public example: buyer-level history is intentionally omitted from docs and landing-page examples."
],
"recommendedOffers": [
{
"endpoint": "company-watch",
"title": "Company Watch",
"path": "/api/company-watch",
"stage": "expansion",
"priceUsd": "$0.20",
"reason": "If the company matters but the blocker is not sharp enough yet, keep the same target warm instead of restarting the trust work later.",
"suggestedUse": "Use this when the next decision depends on meaningful changes over time, not another one-shot read."
},
{
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "Company-level interest often turns into a trust check before a pilot or partnership decision.",
"suggestedUse": "Use this on the same company if the next question is whether it is safe enough to trust."
},
{
"endpoint": "lead-brief",
"title": "Lead Brief",
"path": "/api/lead-brief",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "Once the operator and motion are clear, turn the research into a concrete buyer-facing brief.",
"suggestedUse": "Use this when you want the opening angle and likely objections for outreach."
},
{
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"stage": "entry",
"priceUsd": "$0.05",
"reason": "A low-cost way to pressure-test the strongest public claim before deeper diligence.",
"suggestedUse": "Run this on one claim that would change the buying decision if it proved false."
}
],
"repeatUsePlan": {
"bestNextRoute": "company-watch",
"bestNextTitle": "Company Watch",
"bestNextPath": "/api/company-watch",
"bestNextPriceUsd": "$0.20",
"timing": "now",
"whyNow": "A named-company snapshot is most valuable when it turns into a recurring trust queue instead of going cold until the buyer restarts later.",
"buyAgainWhen": "Buy again when the same company is likely to matter again and you want the next spend triggered by a real change, not by memory.",
"ifUnsure": "If you still are not sure the company deserves deeper spend, return to /api/opportunity-check."
}
},
"evidence": {
"providedSourcesReviewed": [
{
"url": "https://example.com",
"hostname": "example.com",
"fetchedAt": "2026-04-20T18:04:00.000Z",
"retrievalStatus": "skipped",
"contentType": "text/html",
"title": "Synthetic example source",
"description": "Public examples use a stable synthetic host instead of a live fetched page.",
"excerpt": "This example host stands in for a provided source so public docs stay deterministic.",
"note": "Synthetic example host; no live fetch was attempted."
}
],
"verificationReadiness": "A provided source is attached for the company, so the first-pass read can point to a concrete first-party page before manual corroboration.",
"nextChecks": [
"Confirm whether the provided source is first-party and current.",
"Cross-check the strongest fetched claim against one independent source.",
"Treat any missing source pull as an unresolved proof gap."
]
},
"company": {
"canonicalName": "Alpine Foods",
"website": "https://example.com",
"industryFocus": "Food distribution",
"geographyFocus": "United States",
"posture": "Expansion-stage operator balancing category growth with fulfillment credibility.",
"likelyBuyerMotion": "Trust-first commercial motion where proof of supply reliability matters more than novelty."
},
"signals": {
"demandSignals": [
"Hiring for operations and sales",
"Launching new product categories"
],
"trustSignals": [
"Regional specialty food distributor",
"Expanding private-label program"
],
"watchItems": [
"Margin pressure from category expansion",
"Execution strain if demand outpaces operations"
]
},
"decisionCard": {
"verdict": "proceed-carefully",
"headline": "Proceed carefully with Alpine Foods as a real target, but validate one trust-critical proof point before treating the story as durable.",
"recommendedAction": "Treat the company as worth continuing with now, then move straight into the next blocking trust question instead of stopping at a generic snapshot.",
"topReasons": [
"The request already names a company, site, category, geography, claims, and operating signals.",
"The public posture is specific enough to support a useful first-pass decision instead of just loose curiosity.",
"The likely buying motion is trust-first, which means the snapshot can guide the next concrete approval or outreach step."
],
"missingProof": [
"There is still no outside proof yet that category expansion is translating into durable operating quality.",
"The strongest claims still need corroboration beyond the first-party surface."
]
},
"followUpPlan": {
"bestNextMove": {
"endpoint": "company-watch",
"title": "Company Watch",
"path": "/api/company-watch",
"priceUsd": "$0.20",
"reason": "The company matters enough to keep warm, but the next paid step should be triggered by meaningful change instead of another restart from zero."
},
"alternatives": [
{
"trigger": "The next likely blocker is whether the company is safe enough to trust, pilot, or approve for a real partnership motion.",
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"priceUsd": "$0.05",
"reason": "Move directly into a trust or approval read when the blocker is already operational instead of waiting on the watch loop."
},
{
"trigger": "One public claim would change the buying decision if it proved false.",
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"priceUsd": "$0.05",
"reason": "Pressure-test the one claim that matters most before leaning harder into the company story."
},
{
"trigger": "The company still looks promising, but the real next move is outreach instead of approval.",
"endpoint": "lead-brief",
"title": "Lead Brief",
"path": "/api/lead-brief",
"priceUsd": "$0.05",
"reason": "Turn the company read into a concrete outreach angle and objection map."
},
{
"trigger": "Missing proof could materially change a real partnership, procurement, or treasury decision.",
"endpoint": "deep-diligence",
"title": "Deep Diligence",
"path": "/api/deep-diligence",
"priceUsd": "$0.50",
"reason": "Escalate into a deeper verification map when the downside is large enough that a quick snapshot is no longer enough."
}
]
}
}create_company_watch
Keep one company under a live trust watch between decisions
$0.20
Use this when the company already matters but the right next trust move depends on what changes over time, not just one snapshot.
Outcome: The response creates a recurring watch baseline, shows proof gaps and alert triggers, and points to the next justified paid route when the blocker sharpens.
POST /api/company-watch
Example request
{
"companyName": "Alpine Foods",
"website": "https://example.com",
"industry": "Food distribution",
"geography": "United States",
"researchFocus": "growth",
"knownClaims": [
"Regional specialty food distributor",
"Expanding private-label program"
],
"knownSignals": [
"Hiring for operations and sales",
"Launching new product categories"
],
"questions": [
"How concentrated is demand around a few buyers?",
"What trust signals matter most to partners?"
],
"watchGoal": "Keep this named company under a live trust watch until the next approval, outreach, or diligence move is obvious.",
"decisionHorizon": "this-month",
"alertSensitivity": "balanced"
}Example response
{
"requestId": "e98da595-03c8-4c66-9c4c-69a7d4abaf7b",
"createdAt": "2026-04-23T02:20:00.000Z",
"endpoint": "company-watch",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.20",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"companyName": "Alpine Foods",
"website": "https://example.com",
"industry": "Food distribution",
"geography": "United States",
"researchFocus": "growth",
"knownClaims": [
"Regional specialty food distributor",
"Expanding private-label program"
],
"knownSignals": [
"Hiring for operations and sales",
"Launching new product categories"
],
"questions": [
"How concentrated is demand around a few buyers?",
"What trust signals matter most to partners?"
],
"watchGoal": "Keep this named company under a live trust watch until the next approval, outreach, or diligence move is obvious.",
"decisionHorizon": "this-month",
"alertSensitivity": "balanced"
},
"summary": "Company Watch turns one named company into a live trust queue so the desk can keep checking for meaningful changes, keep proof gaps visible, and point to the next justified paid route instead of leaving the work as a one-shot memo.",
"confidence": {
"score": 0.78,
"label": "high",
"rationale": [
"The watch starts with a named company, a first-party site, an explicit watch goal, and concrete operating signals.",
"The initial baseline is specific enough to decide what should trigger a deeper trust move instead of a vague reminder loop."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "provided-source-read",
"evidenceStrength": "strong",
"decisionReadiness": "usable",
"proofBurden": "moderate",
"useFor": [
"Keeping one named company under a live first-pass trust watch between deeper decisions",
"Turning a one-time company read into a recurring queue with clearer escalation rules"
],
"notFor": [
"Treating the watch as final legal, financial, compliance, or investment approval",
"Claiming the company was continuously independently verified outside the supplied source checks"
],
"whyThisIsUseful": [
"It gives the desk a real recurring job instead of waiting for another one-shot request.",
"It makes proof gaps and follow-on routes explicit before the next real decision arrives."
]
},
"decisionSignals": {
"urgency": "active",
"spendReadiness": "core-route",
"proofGapSeverity": "moderate",
"actNowBecause": "The company is already specific enough that ongoing trust movement is more useful than starting from zero again later.",
"holdIf": "Do not treat the watch as final approval. Escalate only when one proof gap, claim, or approval blocker becomes real.",
"escalationTrigger": "Buy again when the watch surfaces a trust, claim, or approval blocker that now justifies source-check, vendor-risk, or deep-diligence.",
"operatorCue": "Use the watch to keep one company warm until the next real decision becomes obvious."
},
"risks": [
"A recurring watch can create false comfort if the strongest proof gap is not revisited when the company narrative changes.",
"Signals from a first-party site can still look healthy even when customer, partner, or control reality is weaker underneath."
],
"opportunities": [
"A recurring watch makes it easier to catch meaningful trust changes before a buyer restarts the same research from zero.",
"The same watch can feed cleaner follow-on purchases into source-check, vendor-risk, or deep-diligence when the blocker sharpens."
],
"nextActions": [
"Keep the named company on a scheduled refresh instead of relying on ad hoc manual reruns.",
"Escalate only when the watch detects a meaningful proof change, a new claim, or a sharper approval blocker.",
"Use the live watch queue to decide whether the next spend should be claim-level, vendor-level, or premium diligence."
],
"questions": [
"What proof gap would most change the next real trust decision on this company?",
"What kind of update should trigger a deeper route instead of staying on the watch?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "Company website",
"url": "https://example.com",
"relevance": "Baseline monitoring surface",
"note": "Use the same first-party source as the recurring watch anchor before you compare later changes."
},
{
"kind": "recommended",
"label": "Independent proof sources",
"url": null,
"relevance": "Follow-on verification",
"note": "Use independent sources when the watch surfaces a change that could alter a real decision."
}
],
"commercial": {
"stage": "expansion",
"buyerRelationship": "new",
"notes": [
"Synthetic public example: buyer-level history is intentionally omitted from docs and public examples.",
"The watch route is a recurring trust lane, so the next step should sharpen the blocker instead of restarting the entire workflow."
],
"recommendedOffers": [
{
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"stage": "entry",
"priceUsd": "$0.05",
"reason": "Use the cheapest follow-on route when one claim becomes the blocker between baseline and action.",
"suggestedUse": "Run this when the watch highlights one shaky public claim or source."
},
{
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "Escalate when the watch says the company is interesting enough that rollout trust or approval now matters.",
"suggestedUse": "Use this when the next blocker is operational or trust-sensitive instead of purely informational."
},
{
"endpoint": "deep-diligence",
"title": "Deep Diligence",
"path": "/api/deep-diligence",
"stage": "expansion",
"priceUsd": "$0.50",
"reason": "Go premium only when the watch proves the missing evidence could change a real procurement, partnership, or treasury decision.",
"suggestedUse": "Use this after the watch isolates a serious proof gap."
}
],
"repeatUsePlan": {
"bestNextRoute": "source-check",
"bestNextTitle": "Source Check",
"bestNextPath": "/api/source-check",
"bestNextPriceUsd": "$0.05",
"timing": "after-proof",
"whyNow": "The watch is most valuable when it narrows the next spend to the single blocker that actually changed.",
"buyAgainWhen": "Buy again when the watch shows a new claim, trust shift, or approval blocker worth pressure-testing.",
"ifUnsure": "If the company still looks too thin for a stronger follow-up, keep it on watch instead of overspending."
}
},
"watch": {
"watchId": "cw_01HZZ9B4X0MMS9Q4S9NXVC1Q0A",
"status": "active",
"cadence": {
"label": "Every 3 hours",
"refreshEveryMinutes": 180,
"nextCheckAt": "2026-04-23T05:20:00.000Z",
"alertSensitivity": "balanced"
},
"decision": "stable",
"whyThisWatchExists": "This watch keeps Alpine Foods under a recurring trust review so the next approval or outreach decision does not start from zero again."
},
"baseline": {
"monitoredCompany": "Alpine Foods",
"website": "https://example.com",
"operatingFocus": "Food distribution",
"decisionHorizon": "this-month",
"monitoringFocus": [
"public positioning changes",
"new trust-sensitive claims",
"signals that justify a deeper route"
],
"trackedSignals": [
"first-party site updates",
"new public claims about growth or trust posture",
"signals that change whether approval or outreach should advance"
],
"proofGaps": [
"Outside proof for the strongest growth and trust claims is still missing.",
"The watch still needs independent corroboration if the decision gets higher-stakes."
],
"whatTriggersAnAlert": [
"A meaningful change in site messaging, trust posture, or operating signals",
"A sharper proof gap that now points to source-check, vendor-risk, or deep-diligence",
"A confidence shift big enough to change the next recommended route"
]
},
"latestRefresh": {
"checkedAt": "2026-04-23T02:20:00.000Z",
"changeState": "baseline-created",
"headline": "Baseline watch created",
"summary": "The watch has a clear first-party anchor, enough structure for a usable recurring baseline, and a moderate proof burden that should stay visible between deeper decisions.",
"confidenceDelta": "new-baseline",
"observedSignals": [
"Named company with a first-party site",
"Existing growth claims and operating signals to monitor over time"
],
"unresolvedRisks": [
"The strongest claims still need outside corroboration before a high-stakes decision relies on them.",
"A change in the public story could still shift the next best route quickly."
],
"notYetVerified": [
"Independent corroboration of the strongest claims",
"Operational proof beyond the first-party surface"
]
},
"queue": {
"nextBestRoute": {
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"priceUsd": "$0.05",
"reason": "Use the cheapest next route when one watch-triggered claim becomes the real blocker."
},
"escalationRoutes": [
{
"trigger": "A trust, rollout, or approval blocker becomes real on the same company.",
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"priceUsd": "$0.05",
"reason": "Escalate when the next question is whether this company is safe enough to trust more deeply."
},
{
"trigger": "Missing proof could now change a real partnership, procurement, or treasury decision.",
"endpoint": "deep-diligence",
"title": "Deep Diligence",
"path": "/api/deep-diligence",
"priceUsd": "$0.50",
"reason": "Use the premium route only when the watch shows the downside is now materially proof-sensitive."
},
{
"trigger": "The signal becomes commercial and the next move is outreach rather than approval.",
"endpoint": "lead-brief",
"title": "Lead Brief",
"path": "/api/lead-brief",
"priceUsd": "$0.05",
"reason": "Turn the watch into a concrete buyer-facing angle when the timing is right."
}
]
},
"activationPlan": {
"firstMilestone": "Create a baseline strong enough that the next route is driven by a real proof gap instead of a vague reminder loop.",
"learningGoal": "Learn whether Alpine Foods is staying stable, getting riskier, or becoming commercial enough to justify the next paid step.",
"returnWhen": "Return to the watch after one meaningful company change, one new public claim, or one new approval blocker.",
"ifNothingChanges": "If nothing meaningful changes, keep the company warm on the watch instead of forcing another one-shot report too early.",
"expectedSecondBuy": {
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"priceUsd": "$0.05",
"reason": "The cheapest strong second buy is pressure-testing the claim or proof point the watch makes most important."
}
},
"evidence": {
"providedSourcesReviewed": [
{
"url": "https://example.com",
"hostname": "example.com",
"fetchedAt": "2026-04-20T18:04:00.000Z",
"retrievalStatus": "skipped",
"contentType": "text/html",
"title": "Synthetic example source",
"description": "Public examples use a stable synthetic host instead of a live fetched page.",
"excerpt": "This example host stands in for a provided source so public docs stay deterministic.",
"note": "Synthetic example host; no live fetch was attempted."
}
],
"verificationReadiness": "The recurring watch starts with a provided company source, which gives the baseline a concrete first-party anchor before deeper verification work happens.",
"nextChecks": [
"Confirm whether the provided source is first-party and current.",
"Cross-check the strongest fetched claim against one independent source.",
"Treat any missing source pull as an unresolved proof gap."
]
}
}assess_vendor_risk
Pressure-test a vendor before pilot approval or procurement
$0.05
Use this when an agent needs a first-pass operational and control review before recommending a rollout.
Outcome: The response scores risk, highlights weak proof areas, and suggests the next diligence questions before approval.
POST /api/vendor-risk
Example request
{
"vendorName": "Northwind AI",
"website": "https://example.com",
"serviceCategory": "AI workflow automation",
"useCase": "Automate intake triage and case routing for an operations team.",
"dataSensitivity": "high",
"criticality": "high",
"geography": "United States",
"knownControls": [
"SOC 2 Type II claimed",
"SSO and role-based access controls"
],
"knownConcerns": [
"No published uptime history",
"Third-party model dependency"
],
"questions": [
"What is the rollback plan if the workflow misroutes tasks?"
]
}Example response
{
"requestId": "5be526a9-1bf2-4a85-b726-08b9011b22f5",
"createdAt": "2026-04-20T18:12:00.000Z",
"endpoint": "vendor-risk",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.05",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"vendorName": "Northwind AI",
"website": "https://example.com",
"serviceCategory": "AI workflow automation",
"useCase": "Automate intake triage and case routing for an operations team.",
"dataSensitivity": "high",
"criticality": "high",
"geography": "United States",
"knownControls": [
"SOC 2 Type II claimed",
"SSO and role-based access controls"
],
"knownConcerns": [
"No published uptime history",
"Third-party model dependency"
],
"questions": [
"What is the rollback plan if the workflow misroutes tasks?"
]
},
"summary": "Northwind AI looks usable for workflow acceleration, but the combination of high-sensitivity data, third-party model dependency, and unclear rollback history pushes the vendor into an elevated-risk bucket.",
"confidence": {
"score": 0.8,
"label": "high",
"rationale": [
"The request includes service category, use case, criticality, data sensitivity, controls, and known concerns.",
"That gives enough detail for a structured first-pass risk posture."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "request-only",
"evidenceStrength": "strong",
"decisionReadiness": "decision-ready",
"proofBurden": "high",
"useFor": [
"Making a first-pass vendor trust call before a pilot, procurement step, or approval review",
"Identifying which control domains need proof before rollout"
],
"notFor": [
"Treating this as legal, compliance, or audit sign-off",
"Assuming claimed controls are proven without reviewing the requested artifacts"
],
"whyThisIsUseful": [
"It ties the risk call to the named workflow instead of giving a vague generic vendor score.",
"It shows which domains are carrying the trust burden so the next diligence request is concrete."
]
},
"decisionSignals": {
"urgency": "active",
"spendReadiness": "premium-route",
"proofGapSeverity": "high",
"actNowBecause": "Vendor approval risk is meaningful enough that unresolved proof gaps can change the rollout decision.",
"holdIf": "Do not pay for another broad route until one control, proof point, or approval blocker is clearly identified.",
"escalationTrigger": "Buy again when one unresolved control or evidence gap still stands between this memo and a real yes-or-no decision.",
"operatorCue": "Keep the follow-up attached to the approval question instead of restarting generic research."
},
"risks": [
"High-sensitivity workflows raise the bar for access controls, auditability, and incident response proof.",
"Third-party model dependency can break reliability assumptions if upstream behavior changes.",
"Missing uptime history weakens confidence for a critical workflow rollout."
],
"opportunities": [
"Claimed SSO and role-based controls suggest the vendor may already support enterprise access patterns.",
"If rollback and human-review controls are strong, the workflow can start with limited operational blast radius."
],
"nextActions": [
"Request security documentation, uptime history, and model-governance details.",
"Test rollback and human-override controls in a pilot environment.",
"Clarify which customer data reaches third-party systems and under what retention terms."
],
"questions": [
"What is the rollback plan if the workflow misroutes tasks?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "Vendor website",
"url": "https://example.com",
"relevance": "Primary vendor source",
"note": "Use the vendor site and trust center as the first proof layer, then compare against buyer evidence."
},
{
"kind": "recommended",
"label": "Security and reliability artifacts",
"url": null,
"relevance": "Control validation",
"note": "Ask for audit reports, uptime data, and architecture documents that match the named controls."
}
],
"commercial": {
"stage": "expansion",
"buyerRelationship": "new",
"notes": [
"Synthetic public example: buyer-level history is intentionally omitted from docs and landing-page examples."
],
"recommendedOffers": [
{
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"stage": "entry",
"priceUsd": "$0.05",
"reason": "A cheap claim check can validate the one proof point that matters most before a final yes or no.",
"suggestedUse": "Use this on the strongest vendor claim that still feels unproven."
},
{
"endpoint": "lead-brief",
"title": "Lead Brief",
"path": "/api/lead-brief",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "Once the risk picture is clear, the next step is arming the operator-facing conversation.",
"suggestedUse": "Use this if you need a tight buyer-specific follow-up after diligence clears."
},
{
"endpoint": "company-intel",
"title": "Company Intel",
"path": "/api/company-intel",
"stage": "core",
"priceUsd": "$0.05",
"reason": "Risk posture is easier to interpret when paired with a view of how the company actually sells and operates.",
"suggestedUse": "Use this if you still need a cleaner view of the company itself."
}
],
"repeatUsePlan": {
"bestNextRoute": "source-check",
"bestNextTitle": "Source Check",
"bestNextPath": "/api/source-check",
"bestNextPriceUsd": "$0.05",
"timing": "after-proof",
"whyNow": "After a risk review, the most efficient next move is usually settling one unresolved proof point.",
"buyAgainWhen": "Buy again when one claim, control, or artifact now stands between the current risk memo and a real yes-or-no decision.",
"ifUnsure": "If the blocker is still broad instead of specific, go to /api/company-intel before buying another trust route."
}
},
"rating": {
"level": "elevated",
"score": 72,
"rationale": "Controls appear plausible, but criticality and data sensitivity are high relative to the unverified operational proof."
},
"domains": {
"security": "Moderate. Claimed enterprise controls help, but proof of enforcement is still needed.",
"compliance": "Elevated. Sensitive workflows can trigger policy, retention, and third-party review requirements.",
"operational": "High. Critical workflow automation needs rollback and exception handling before broad rollout.",
"financial": "Moderate. The main risk is failed deployment or vendor dependence rather than immediate balance-sheet exposure.",
"dependency": "High. Model and infrastructure dependency need explicit fallback plans."
}
}create_deep_diligence_report
Build a Deep Diligence Plan before serious spend or approval
$0.50
Use this when an agent needs more than a quick screen and wants a structured proof map before a meaningful trust decision.
Outcome: The response separates what was provided from what was not verified, then returns trust signals, risk flags, missing evidence, recommended verification sources, and next actions.
POST /api/deep-diligence
Example request
{
"subject": "Northwind AI",
"subjectType": "vendor",
"website": "https://example.com",
"geography": "United States",
"industry": "AI workflow automation",
"objective": "Decide whether this vendor deserves a serious trust review before procurement or pilot planning.",
"knownClaims": [
"SOC 2 Type II claimed",
"Enterprise-ready deployment posture"
],
"knownSignals": [
"No public uptime history",
"Third-party model dependency",
"Growing enterprise go-to-market motion"
],
"questions": [
"Which missing proof would most change the trust decision for this vendor?"
]
}Example response
{
"requestId": "9183d1b7-f6f8-41eb-a6d6-6c74eb2d3af3",
"createdAt": "2026-04-20T18:13:00.000Z",
"endpoint": "deep-diligence",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.50",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"subject": "Northwind AI",
"subjectType": "vendor",
"website": "https://example.com",
"geography": "United States",
"industry": "AI workflow automation",
"objective": "Decide whether this vendor deserves a serious trust review before procurement or pilot planning.",
"knownClaims": [
"SOC 2 Type II claimed",
"Enterprise-ready deployment posture"
],
"knownSignals": [
"No public uptime history",
"Third-party model dependency",
"Growing enterprise go-to-market motion"
],
"questions": [
"Which missing proof would most change the trust decision for this vendor?"
]
},
"summary": "Northwind AI needs a structured decision-prep map because the trust decision still depends on proof gaps, missing evidence, and verification steps that were not resolved by the supplied information alone.",
"confidence": {
"score": 0.83,
"label": "high",
"rationale": [
"The request names the subject type, website, geography, industry, objective, claims, and signals.",
"That gives enough structure for a serious first-pass diligence map rather than a generic summary."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "provided-source-read",
"evidenceStrength": "strong",
"decisionReadiness": "decision-ready",
"proofBurden": "high",
"useFor": [
"Preparing a serious trust decision where missing proof could change the outcome",
"Building a verification map another agent or human can execute immediately"
],
"notFor": [
"Treating this as final legal, financial, compliance, or investment diligence",
"Claiming the recommended sources were already checked unless they were supplied directly and fetched"
],
"whyThisIsUseful": [
"A provided source is attached, so the plan starts from a real proof surface instead of only abstract inputs.",
"It shows exactly which evidence gaps still matter before a serious decision should move forward."
]
},
"decisionSignals": {
"urgency": "immediate",
"spendReadiness": "premium-route",
"proofGapSeverity": "high",
"actNowBecause": "This is already a premium proof-gap map for a serious trust decision where missing evidence can materially change the outcome.",
"holdIf": "Do not treat the plan as final approval until the strongest recommended verification source is actually checked.",
"escalationTrigger": "Buy again only after the premium map isolates the one blocker that still decides the outcome.",
"operatorCue": "Use the premium map to isolate the single blocker that still matters instead of widening the work again."
},
"risks": [
"Enterprise-ready positioning can outrun the actual maturity of control evidence or rollback planning.",
"Third-party model dependency may create reliability and governance risk that marketing language will not surface cleanly.",
"No public uptime history weakens confidence in the operating story."
],
"opportunities": [
"The subject is specific enough to turn into a concrete proof checklist instead of a vague diligence memo.",
"Named claims and signals make it easier to focus manual verification on the handful of facts that matter most."
],
"nextActions": [
"Request proof for the claimed controls and current uptime or incident history.",
"Validate how much customer data or operational dependency reaches third-party systems.",
"Treat missing rollback, governance, or audit evidence as blockers until they are narrowed."
],
"questions": [
"Which missing proof would most change the trust decision for this vendor?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "First-party source",
"url": "https://example.com",
"relevance": "Primary proof surface",
"note": "This source was provided in the request. It is a starting point for review, not proof that the claims were independently verified."
},
{
"kind": "recommended",
"label": "Independent corroboration set",
"url": null,
"relevance": "Trust validation",
"note": "These are recommended next sources to check. They were not retrieved or independently verified by this example response."
}
],
"commercial": {
"stage": "expansion",
"buyerRelationship": "new",
"notes": [
"Synthetic public example: buyer-level history is intentionally omitted from docs and landing-page examples.",
"This is the premium trust product, so the next offer should narrow the final blocking question instead of restarting the research."
],
"recommendedOffers": [
{
"endpoint": "vendor-risk",
"title": "Vendor Risk",
"path": "/api/vendor-risk",
"stage": "expansion",
"priceUsd": "$0.05",
"reason": "A structured risk snapshot is the fastest follow-up when deep diligence surfaces a control or rollout concern that still needs clearer framing.",
"suggestedUse": "Use this when the buyer now needs a concise risk memo for approval or operator review."
},
{
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"stage": "entry",
"priceUsd": "$0.05",
"reason": "A claim check is the cheapest way to settle the single proof point that still feels shaky after the larger review.",
"suggestedUse": "Run this on the one claim or source that most affects the decision."
},
{
"endpoint": "company-intel",
"title": "Company Intel",
"path": "/api/company-intel",
"stage": "core",
"priceUsd": "$0.05",
"reason": "A broader company view helps explain how the diligence findings map back to the subject's operating posture.",
"suggestedUse": "Use this if you need a cleaner operating snapshot after the premium diligence pass."
}
],
"repeatUsePlan": {
"bestNextRoute": "vendor-risk",
"bestNextTitle": "Vendor Risk",
"bestNextPath": "/api/vendor-risk",
"bestNextPriceUsd": "$0.05",
"timing": "after-action",
"whyNow": "The premium map becomes more usable when it turns into one tight follow-up memo for approval or operator review.",
"buyAgainWhen": "Buy again after the diligence plan exposes the one operating or control blocker that still needs a sharper writeup.",
"ifUnsure": "If the premium review still feels too broad, use /api/source-check on the single proof point that matters most."
}
},
"evidence": {
"providedSourcesReviewed": [
{
"url": "https://example.com",
"hostname": "example.com",
"fetchedAt": "2026-04-20T18:04:00.000Z",
"retrievalStatus": "skipped",
"contentType": "text/html",
"title": "Synthetic example source",
"description": "Public examples use a stable synthetic host instead of a live fetched page.",
"excerpt": "This example host stands in for a provided source so public docs stay deterministic.",
"note": "Synthetic example host; no live fetch was attempted."
}
],
"verificationReadiness": "The first-party source is attached in the response shape, but the premium plan still frames the result as decision prep rather than a final verified verdict.",
"nextChecks": [
"Confirm whether the provided source is first-party and current.",
"Cross-check the strongest fetched claim against one independent source.",
"Treat any missing source pull as an unresolved proof gap."
]
},
"report": {
"executiveSummary": "Northwind AI may deserve serious diligence, but the trust decision should stay open until control evidence, uptime proof, and dependency handling are checked directly.",
"subjectProfile": {
"name": "Northwind AI",
"type": "vendor",
"website": "https://example.com",
"geography": "United States",
"industry": "AI workflow automation",
"operatingContext": "Vendor-level diligence where control evidence, reliability, and delivery risk matter before approval or pilot."
},
"trustSignals": [
"SOC 2 Type II claimed",
"Enterprise-ready deployment posture",
"Growing enterprise go-to-market motion",
"A first-party website or source is available for direct verification.",
"The diligence objective is clearly stated, which narrows the verification path."
],
"riskFlags": [
"Important claims are present, but the request does not prove them on its own.",
"Signals may indicate momentum, but they do not confirm durable trust or operating quality.",
"Operational or control gaps can stay hidden until a pilot or incident forces them into view.",
"Missing evidence can make a subject appear stronger than it really is if the buyer relies on positioning alone."
],
"missingEvidence": [
"Independent proof for the named claims",
"Security, uptime, and rollback evidence",
"Time-bounded proof that the most important signals are current"
],
"whatWasProvided": [
"Subject type, website, geography, industry, objective, known claims, and known signals were supplied by the buyer.",
"The response used the provided website as the first-party surface to review later."
],
"whatWasNotVerified": [
"The claimed controls and deployment posture were not independently verified by this response.",
"No audit report, uptime history, or customer corroboration was retrieved in this example."
],
"recommendedVerificationSources": [
"Security documentation, uptime history, and trust center materials",
"Customer references and implementation evidence",
"Official first-party materials, docs, or trust pages",
"Independent references such as customer evidence, audits, interviews, or third-party reporting",
"Evidence that ties claims to current operations instead of old or promotional material"
],
"comparisonAngles": [
"Compare trust-center claims against real uptime, incident, or rollback evidence.",
"Compare marketing-level enterprise readiness claims against the actual control artifacts available.",
"Compare the proof burden for this vendor against the blast radius of the intended workflow."
]
}
}create_lead_brief
Qualify a lead and shape the first outreach angle
$0.05
Use this when an agent wants to decide whether a prospect is worth attention and what message frame is most likely to land.
Outcome: The response returns fit, likely pains, objections, buying signals, and next actions for sales or operator workflows.
POST /api/lead-brief
Example request
{
"companyName": "Harbor Ops",
"website": "https://example.com",
"contactName": "Avery Stone",
"role": "VP of Operations",
"industry": "Maritime logistics",
"geography": "United States",
"painHypothesis": "Manual vessel-status updates create avoidable coordination delays.",
"offer": "Agentic operations desk for status monitoring and follow-up drafting.",
"knownSignals": [
"Hiring coordinators",
"Publicly discussing workflow automation",
"Expanding into new ports"
],
"questions": [
"What business trigger makes this lead worth prioritizing this quarter?"
]
}Example response
{
"requestId": "0d1f8aef-6d1a-4bd8-8dd6-c6058a11333b",
"createdAt": "2026-04-20T18:14:00.000Z",
"endpoint": "lead-brief",
"service": {
"name": "Agent Research Desk",
"version": "1.7.1",
"x402Enabled": true,
"network": "eip155:8453",
"price": "$0.05",
"url": "https://agent-research-brief-api-production.up.railway.app"
},
"input": {
"companyName": "Harbor Ops",
"website": "https://example.com",
"contactName": "Avery Stone",
"role": "VP of Operations",
"industry": "Maritime logistics",
"geography": "United States",
"painHypothesis": "Manual vessel-status updates create avoidable coordination delays.",
"offer": "Agentic operations desk for status monitoring and follow-up drafting.",
"knownSignals": [
"Hiring coordinators",
"Publicly discussing workflow automation",
"Expanding into new ports"
],
"questions": [
"What business trigger makes this lead worth prioritizing this quarter?"
]
},
"summary": "Harbor Ops looks like a strong operational-efficiency lead because the company is expanding while public signals suggest workflow strain, and the target contact likely owns the pain directly.",
"confidence": {
"score": 0.78,
"label": "high",
"rationale": [
"The request names a company, contact role, pain hypothesis, offer, and several commercial signals.",
"That gives enough structure for an account-specific outreach hypothesis."
]
},
"trustProfile": {
"operatingMode": "first-pass",
"verificationStatus": "request-only",
"evidenceStrength": "strong",
"decisionReadiness": "decision-ready",
"proofBurden": "moderate",
"useFor": [
"Qualifying whether a named account deserves outreach now",
"Shaping a first message around evidence-backed pain instead of a generic pitch"
],
"notFor": [
"Treating this as a final account plan or verified buying-intent proof",
"Assuming the named pain hypothesis is true without checking fresh account signals"
],
"whyThisIsUseful": [
"It turns scattered account signals into a concrete outreach angle and objection map.",
"It stays honest about proof gaps, which makes the output safer for real operator use."
]
},
"decisionSignals": {
"urgency": "active",
"spendReadiness": "core-route",
"proofGapSeverity": "moderate",
"actNowBecause": "The lead brief is most useful while the account is still warm and the first message has not been sent yet.",
"holdIf": "Do not buy another outreach-focused route if the real blocker has become trust, proof, or approval risk.",
"escalationTrigger": "Buy again when outreach or qualification makes the same account the clear next diligence target.",
"operatorCue": "Turn the outreach signal into account trust work while the target is still warm."
},
"risks": [
"The pain hypothesis may be directionally right but still low-priority if there is no near-term service disruption.",
"Operational leaders may resist automation claims unless the rollout path is low-friction and reversible."
],
"opportunities": [
"Expansion into new ports can magnify coordination pain and make workflow automation newly urgent.",
"A VP of Operations is likely close enough to the process to validate pain quickly."
],
"nextActions": [
"Verify whether the company is actively scaling the exact workflow named in the hypothesis.",
"Tailor the first message around delay reduction, not generic AI productivity.",
"Prepare one low-risk pilot example that fits a VP of Operations decision frame."
],
"questions": [
"What business trigger makes this lead worth prioritizing this quarter?"
],
"sourceAttribution": [
{
"kind": "provided",
"label": "Company website",
"url": "https://example.com",
"relevance": "Account framing",
"note": "Use the company site to confirm operating model, expansion narrative, and contact relevance."
},
{
"kind": "recommended",
"label": "Public growth and workflow signals",
"url": null,
"relevance": "Prioritization",
"note": "Validate hiring, expansion, and process-change signals before outreach."
}
],
"commercial": {
"stage": "expansion",
"buyerRelationship": "new",
"notes": [
"Synthetic public example: buyer-level history is intentionally omitted from docs and landing-page examples.",
"This is already close to a sales action, so the next product should strengthen trust or sharpen one claim before outreach."
],
"recommendedOffers": [
{
"endpoint": "company-intel",
"title": "Company Intel",
"path": "/api/company-intel",
"stage": "core",
"priceUsd": "$0.05",
"reason": "A lead brief is stronger when paired with a fuller read on the account and its operating posture.",
"suggestedUse": "Use this if you want a cleaner account map before sending the first message."
},
{
"endpoint": "source-check",
"title": "Source Check",
"path": "/api/source-check",
"stage": "entry",
"priceUsd": "$0.05",
"reason": "Validate one high-impact claim before you use it in outreach.",
"suggestedUse": "Run this on the proof point you plan to cite to the buyer."
},
{
"endpoint": "research-brief",
"title": "Research Brief",
"path": "/api/research-brief",
"stage": "core",
"priceUsd": "$0.05",
"reason": "If the account points to a bigger market theme, expand it into a broader research plan.",
"suggestedUse": "Use this when the account-level insight should turn into a repeatable market thesis."
}
],
"repeatUsePlan": {
"bestNextRoute": "company-intel",
"bestNextTitle": "Company Intel",
"bestNextPath": "/api/company-intel",
"bestNextPriceUsd": "$0.05",
"timing": "after-action",
"whyNow": "A lead brief is strongest when it quickly turns into a clearer account-level trust picture.",
"buyAgainWhen": "Buy again once outreach, qualification, or account review makes the same company the clear next diligence target.",
"ifUnsure": "If you only need to validate one talking point before outreach, use /api/source-check instead."
}
},
"lead": {
"companyName": "Harbor Ops",
"contactName": "Avery Stone",
"role": "VP of Operations",
"industry": "Maritime logistics",
"geography": "United States",
"fit": "strong"
},
"brief": {
"openingAngle": "Frame the outreach around avoidable coordination delays during expansion, not around AI as a category.",
"valueHypothesis": "If the team is scaling across more ports, a monitoring and follow-up desk could reduce manual status chasing and escalation lag.",
"likelyPainPoints": [
"Manual status updates across fragmented systems",
"Escalation delays when coordinators juggle too many exceptions",
"Growth forcing more work through the same operating headcount"
],
"objectionWatchlist": [
"Fear of workflow disruption",
"Need for clear human override",
"Skepticism about data accuracy or integration effort"
],
"buyingSignals": [
"Hiring coordinators",
"Publicly discussing workflow automation",
"Expanding into new ports"
]
}
}Capabilities
Available Endpoints
Full schemas and machine-readable examples for every route are published at /docs.json.
check_opportunity
Opportunity Check
$0.01
Best first buy when an agent is unsure whether a target deserves more diligence. Returns a lightweight pass/watch/skip screen, the trust question, the fastest proof path, and the best next paid product.
POST /api/opportunity-check
Use this when: Use this when you are unsure whether a target or claim deserves more paid work.
Do not use this when: Do not use this when you already know the blocker and need a company snapshot, vendor review, claim check, lead brief, or premium diligence map.
Expected input: subject
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
create_research_brief
Research Brief
$0.05
Best when the buyer needs a structured first-pass memo before deciding what to do next. Returns executive framing, working hypotheses, research questions, search queries, interview targets, risks, opportunities, and next actions.
POST /api/research-brief
Use this when: Use this when the buyer needs general structured research and the next step is still open-ended.
Do not use this when: Do not use this when one company, one vendor, one claim, or one lead is already the clear focus.
Expected input: subject, objective
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
create_company_intel
Company Intel
$0.05
Best when the buyer is screening one account, partner, or target company. Returns a decision card, posture, signals, risks, opportunities, follow-up questions, and the clearest next trust route for the same company.
POST /api/company-intel
Use this when: Use this for a quick company snapshot before outreach, approval, or a deeper trust review.
Do not use this when: Do not use this when the blocker is a single claim, a vendor-control question, or a decision that hinges on missing proof.
Expected input: companyName
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
create_company_watch
Company Watch
$0.20
Best when the buyer wants the desk to keep one company warm between decisions instead of restarting from zero later. Returns a live watch baseline, refresh cadence, proof gaps, alert triggers, and the next justified paid route when the blocker sharpens.
POST /api/company-watch
Use this when: Use this when one named company matters enough to keep monitoring between outreach, approval, partnership, or trust decisions.
Do not use this when: Do not use this when you only need a one-time company snapshot or when one single claim is already the only blocker.
Expected input: companyName, watchGoal
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
check_source_claim
Source Check
$0.05
Best low-cost follow-up when one proof point still feels shaky. Returns a claim verdict, source-quality notes, corroboration steps, risks, opportunities, and open questions.
POST /api/source-check
Use this when: Use this when one claim or source is the blocker and you want the cheapest focused check first.
Do not use this when: Do not use this when you need a broader company, vendor, lead, or general research view.
Expected input: claim
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
assess_vendor_risk
Vendor Risk
$0.05
Best when the buyer already knows the vendor and needs a concise trust memo. Returns a risk level, domain-by-domain notes, opportunities, next actions, and diligence questions.
POST /api/vendor-risk
Use this when: Use this before a pilot, procurement review, partnership decision, or vendor approval.
Do not use this when: Do not use this when you only need a cheap initial screen or when one source claim is the only blocker.
Expected input: vendorName, serviceCategory, useCase
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
create_deep_diligence_report
Deep Diligence
$0.50
Best when the buyer needs a decision-prep map before procurement, partnership, or another serious trust decision. Returns an executive summary, subject profile, trust signals, risk flags, what was provided, what was not verified, missing evidence, recommended verification sources, and next actions based on the supplied information.
POST /api/deep-diligence
Use this when: Use this when missing evidence could change a real decision and the buyer needs a structured verification map.
Do not use this when: Do not use this as legal diligence, investment diligence, compliance approval, or a claim that external facts were independently verified.
Expected input: subject, subjectType
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
create_lead_brief
Lead Brief
$0.05
Best when the buyer already has a target account and wants a sharper first message. Returns fit assessment, opening angle, likely pains, objections, buying signals, risks, opportunities, and next actions.
POST /api/lead-brief
Use this when: Use this before outreach when the buyer wants a tighter account view and a clearer first message.
Do not use this when: Do not use this when the main question is vendor approval, claim verification, or broader market research.
Expected input: companyName
Returns: structured JSON with summary, confidence, trustProfile, decisionSignals, risks, opportunities, nextActions, questions, sourceAttribution, commercial.
Trust & Safety
Use the output responsibly
- This service does not provide financial, legal, tax, compliance, or investment advice.
- Outputs are first-pass intelligence meant to speed up research, not replace human verification.
- Buyers should verify important claims, counterparties, vendors, and operational assumptions before making high-stakes decisions.
- Deep Diligence is a decision-prep report, not a final legal, financial, compliance, or investment opinion.
Enterprise Path
Higher-touch rollout
Custom pricing based on route mix, usage shape, and support level.
- Start with the live public routes so the use case is proven cheaply.
- Pick the one or two high-value diligence flows that should become repeatable.
- Move those flows into a private rollout plan with custom guidance, support, and pricing.
Changelog
1.8.0
- Added a dedicated /agents front door and /agent-capabilities.json manifest so other agents can discover the cleanest entry path without reading the full human-facing site first.
- Connected the public agent-facing surfaces to live route emphasis, current scout focus, and the repeat Company Watch loop while keeping all public telemetry aggregate-only.
- Expanded the public and private dashboards with agent-entry traffic so overnight monitoring can show whether agent-native acquisition is actually starting to work.
The original launch proof is still valid, and the current launch pass keeps the working payment setup while tightening privacy, positioning, and deep-diligence boundaries.