CrewAI Tools for Brazil CVM Filings
Building CrewAI CVM filings agents? Expose apicvm as tools: resolve company, list documents, download file. Crews should cite document UUIDs. This is not a hosted CrewAI integration.
The problem
Multi-agent research crews that mention Brazilian issuers need primary filings, not blog summaries. Portal HTML is a poor tool surface. HTTPS JSON with Bearer auth fits tool-calling loops.
Tool sketch
import os, requests
from crewai.tools import tool
BASE = os.environ["APICVM_URL"]
H = {"Authorization": f"Bearer {os.environ['APICVM_KEY']}"}
@tool("List Brazil CVM filings")
def list_cvm_filings(ticker: str, doc_type: str = "DFP", year: int = 2024) -> str:
"""List CVM documents for a B3 ticker via apicvm."""
r = requests.get(
f"{BASE}/v1/documents",
headers=H,
params={"ticker": ticker, "type": doc_type, "year": year, "perPage": 20},
timeout=60,
)
r.raise_for_status()
rows = r.json()["data"]
return "\n".join(f"{d['id']}\t{d['name']}" for d in rows)Crew roles that fit
- Researcher: list FRE risk sections (
namefilter) for a ticker - Archivist: download PDF bytes to object storage by UUID
- Analyst: wait for async extraction callbacks before quoting page text
Do not invent a sync "get full FRE text" tool — extraction is async via callback (Pro).
Current limitations
- No push notifications for new filings; crews must re-list
- Rate limits apply; share one key carefully across parallel agents
- Empty lists can be corpus gaps
Next steps
Ready to integrate?
Get an API key and start querying Brazilian CVM filings programmatically.