Build a Sales Research Agent with AIsa

The problem

Sales and partnerships teams spend hours building target lists, identifying the right contacts, and finding timely reasons to reach out. An agent can handle the research, but enrichment incurs costs, returns personal data, and CRM or outreach actions carry real-world side effects. A good outcome is a concise, scored list of accounts and contacts with source links and trigger events, plus draft messaging, with nothing written or sent until you approve.

Prompt

Give this to your agent once AIsa is connected via MCP:

Use AIsa to find <N> target accounts matching: <industry>, <headcount range>, <region>, <technology or signal>.
Find the right AIsa tools with `search` and quote paid calls with `batch_quote`. Ask me before any paid call, and especially before enrichment.
Step 1: search companies and people by criteria (no enrichment yet). Step 2: pull recent news for each account to find a trigger event. Step 3: score each account against my criteria and show the scoring.
Only after I approve a shortlist, enrich those people, requesting only name, title, and work email.
Return a table: account | why it fits | trigger event (with link and date) | contact role | missing or uncertain fields. Then a 3-sentence draft opener per account.
Do not create contacts, add anyone to a sequence, or send email.

APIs and Skills used

Through MCP, your agent reaches these via a few meta tools at https://mcp.aisa.one/mcp: search, list_categories, get_details, and batch_quote are free; batch_use (up to 20 calls) is billed per call.

Step	What the agent calls	Reference
Build a target list	Company search by headcount, industry, location, technology	Organization Search
Find people	People search (contact details withheld)	People API Search
Trigger events	Company news by category and date	News Articles Search
Company context	Website traffic snapshot	Website Traffic Snapshot, Use Similarweb
Enrich the shortlist	One company by domain; one person by identifiers	Organization Enrichment, People Enrichment
Creator targets	Creator discovery for influencer programs	KOL Creator Discovery
Write actions (gated)	Create a contact; add to a sequence	Create a Contact, Add Contacts to a Sequence

People search indicates whether a match exists and flags if an email or phone number is available; enrichment reveals the details and is where the enrichment charge applies. Revealing personal emails or phone numbers is opt-in and billed extra.

Example output

Illustrative only. Companies, roles, and links are placeholders, not real records.

Account	Why it fits	Trigger event	Contact role	Missing / uncertain
Example Corp	200–500 staff, EU, uses ExampleDB	Opened a new office, 2026-01-08 (link)	Head of Data	Work email not yet enriched
Sample Labs Inc.	50–200 staff, EU, hiring data roles	Funding announcement, 2025-12-20 (link)	VP Engineering	Headcount source is a range

Score: fit criteria matched (4/5, 3/5), shown per account. Draft opener (Example Corp): “Congrats on the new office…”

Guardrails

Read first, enrich later. Discover broadly, then enrich only the approved shortlist. Do not enrich every result by default.
Data minimization. Request only the fields the workflow requires; record the source and retrieval time; and comply with relevant privacy, consent, and outreach rules.
Inspectable scoring. Keep source data, derived scores, model interpretation, and missing fields separate. The model must not invent revenue, headcount, titles, contact details, or buying intent.
Drafting is not sending. Before creating a contact, adding someone to a sequence, sending a message, or updating a CRM: verify the connected account and permissions, show the exact action and target, obtain explicit approval, execute once, and verify the result. Read access does not imply write permission.
Quote, then ask. Get a free quote with batch_quote and run the same calls with batch_use only after approval. A quote checks the estimate, not the final charge. See Billing & Credits.
When AIsa may not be necessary: A direct CRM or data-provider integration is simpler if you use a single provider and already have stable enrichment, consent, and outreach processes.

Social Intelligence Agent

Creator and social signals.

SEO Keyword Research

Competitor and keyword gaps.

Authentication

Keys and OAuth.

Security & Privacy

Agent safety controls.

​The problem

​Prompt

​APIs and Skills used

​Example output

​Guardrails

​Related

Social Intelligence Agent

SEO Keyword Research

Authentication

Security & Privacy

The problem

Prompt

APIs and Skills used

Example output

Guardrails

Related