The Problem
Meeting transcripts contain decisions, action items, and context that AI agents need to reference later. But transcripts sit in Zoom recordings or Google Docs, disconnected from the agent's knowledge base. When a user asks their agent 'what did we decide about the pricing model?', the agent has no access to meeting context. Manually copying meeting notes into agent memory is tedious and inconsistent.
The Scavio Solution
Build a pipeline that takes meeting transcripts, extracts key decisions and action items, enriches them with current web context from Scavio (e.g., searching for referenced products, competitors, or pricing), and stores structured summaries in the agent's knowledge base. Scavio provides the real-time web context that turns raw transcript text into actionable, enriched knowledge entries.
Before
Before the pipeline, meeting decisions lived in unstructured transcripts. The AI agent could not reference meeting context. Users re-explained decisions to the agent repeatedly, losing time and creating inconsistencies.
After
After deploying the pipeline, each meeting produces structured knowledge entries enriched with current web data. The agent references meeting decisions naturally. Users ask 'what did we decide about X?' and get accurate answers with links to supporting data.
Who It Is For
Teams building AI agents that need access to meeting context. Product managers and engineering leads who want meeting decisions searchable by their AI assistant.
Key Benefits
- Meeting decisions auto-extracted and stored in agent knowledge base
- Web context enrichment adds current data to meeting references
- Agent answers questions about past meetings with cited sources
- Structured entries replace unstructured transcript dumps
- Pipeline cost under $0.10 per meeting for web enrichment queries
Python Example
import requests
import json
from datetime import datetime
API_KEY = "your_scavio_api_key"
def enrich_meeting_topic(topic: str) -> dict:
"""Enrich a meeting topic with current web context."""
res = requests.post(
"https://api.scavio.dev/api/v1/search",
headers={"x-api-key": API_KEY},
json={"platform": "google", "query": topic, "ai_overview": True},
timeout=15,
)
res.raise_for_status()
data = res.json()
return {
"topic": topic,
"ai_summary": data.get("ai_overview", {}).get("text", ""),
"top_sources": [{"title": r.get("title", ""), "link": r.get("link", ""), "snippet": r.get("snippet", "")} for r in data.get("organic", [])[:3]],
"enriched_at": datetime.utcnow().isoformat(),
}
def process_meeting(decisions: list[dict]) -> list[dict]:
"""Enrich meeting decisions with web context."""
enriched = []
for decision in decisions:
context = enrich_meeting_topic(decision["topic"])
enriched.append({**decision, "web_context": context})
return enriched
decisions = [
{"topic": "Scavio API pricing vs SerpAPI", "decision": "Switch to Scavio for cost savings"},
{"topic": "n8n vs Make automation platform", "decision": "Evaluate both with pilot workflows"},
]
enriched = process_meeting(decisions)
for d in enriched:
print(f"Decision: {d['decision']}")
print(f" Context: {d['web_context']['ai_summary'][:100]}...")JavaScript Example
const API_KEY = "your_scavio_api_key";
async function enrichTopic(topic) {
const res = await fetch("https://api.scavio.dev/api/v1/search", {
method: "POST",
headers: { "x-api-key": API_KEY, "content-type": "application/json" },
body: JSON.stringify({ platform: "google", query: topic, ai_overview: true }),
});
const data = await res.json();
return {
topic,
aiSummary: data.ai_overview?.text ?? "",
sources: (data.organic ?? []).slice(0, 3).map((r) => ({ title: r.title ?? "", link: r.link ?? "" })),
};
}
const decisions = [
{ topic: "Scavio API pricing vs SerpAPI", decision: "Switch to Scavio" },
{ topic: "n8n vs Make automation", decision: "Pilot both" },
];
for (const d of decisions) {
const ctx = await enrichTopic(d.topic);
console.log(`${d.decision}: ${ctx.aiSummary.slice(0, 100)}...`);
}Platforms Used
Web search with knowledge graph, PAA, and AI overviews