EU funding calls from the Funding & Tenders Portal

Reads the public search API that the portal web page itself uses. It needs no account. The Commission legal notice (https://commission.europa.eu/legal-notice_en) says: "This means that reuse is allowed, provided appropriate credit is given and changes are indicated." The licence is Creative Commons Attribution 4.0 International (CC BY 4.0) for content owned by the EU, unless otherwise indicated. Credit the source as: "Source: European Commission, Funding & Tenders Portal, CC BY 4.0".

The API is not documented for third parties and the public key in the URL (apiKey=SEDIA) is the one the portal page sends. It can change without notice. Make a few targeted requests, not a crawl.

Author: Don Mangu (GitHub). This skill needs no token. For scheduled runs, cleaned text, a CSV or JSON export and a full read of all open calls, the author also runs a paid hosted version on the Apify Store: EU Funding Calls API (disclosure: the author owns it). You never need it for the steps below. This is an unofficial tool and is not affiliated with, endorsed by or operated by the European Commission.

Example prompts

• "Which Horizon Europe topics are open and close in the next month?"
• "List open Digital Europe calls with a budget above 10 million euros."
• "What is the deadline and budget of HORIZON-CL5-2027-07-D3-25?"

Out of scope: "Should we apply to this call?" The skill lists calls. It gives no advice on eligibility of a given organisation.

Step 1: know the codes

• Status ids: 31094501 forthcoming, 31094502 open, 31094503 closed.
• Record type: 1 topic (the normal grant topic), 2 EuropeAid list call, 8 competitive call of a funded project.
• frameworkProgramme ids: 43108390 Horizon Europe, 43152860 Digital Europe, 43252405 LIFE, 43353764 Erasmus+, 43251589 CERV, 44181033 European Defence Fund, 43251567 Connecting Europe Facility, 43252476 Single Market Programme, 43298916 Euratom.

Step 2: send the request

POST https://api.tech.ec.europa.eu/search-api/prod/rest/search?apiKey=SEDIA&text=*&pageSize=100&pageNumber=1 with three multipart parts named query, languages and sort. Each part must carry the content type application/json. Without it the API answers HTTP 500. A page holds at most 100 records and the API refuses results beyond 10,000 per query. The query is an Elasticsearch-style filter: terms on type, status and frameworkProgramme, and range on deadlineDate or ccm2Id.

Each result has metadata with lists as values: identifier, title, status, typesOfAction, deadlineDate, startDate, descriptionByte (HTML), topicConditions (HTML), budgetOverview (JSON text) and actions (JSON text with plannedOpeningDate and deadlineDates). Use the dates in actions for the cut-offs of a topic: the deadlineDate field can lag behind for topics with several cut-offs.

Step 3: run it

import datetime, json, time, urllib.request, uuid

URL = "https://api.tech.ec.europa.eu/search-api/prod/rest/search?apiKey=SEDIA&text=*&pageSize=100&pageNumber=1"
UA = "MyResearchTool/1.0 (+https://example.org/contact)"   # name your tool and give your own contact
OPEN, FORTHCOMING = "31094502", "31094501"
TODAY = datetime.date.today().isoformat()

def post(query, sort):
    """POST the three JSON parts (query, languages, sort) as multipart/form-data."""
    boundary = uuid.uuid4().hex
    parts = {"query": query, "languages": ["en"], "sort": sort}
    body = b""
    for name, value in parts.items():
        body += (f'--{boundary}\r\nContent-Disposition: form-data; name="{name}"; filename="{name}.json"\r\n'
                 f'Content-Type: application/json\r\n\r\n{json.dumps(value)}\r\n').encode()
    body += f"--{boundary}--\r\n".encode()
    req = urllib.request.Request(URL, data=body, headers={
        "User-Agent": UA, "Content-Type": f"multipart/form-data; boundary={boundary}"})
    time.sleep(0.5)                                        # at most 2 requests per second
    return json.load(urllib.request.urlopen(req, timeout=60))

query = {"bool": {"must": [{"terms": {"type": ["1"]}}, {"terms": {"status": [OPEN]}},
                           {"terms": {"frameworkProgramme": ["43108390"]}},   # 43108390 = Horizon Europe
                           {"range": {"deadlineDate": {"gte": TODAY + "T00:00:00.000+0000"}}}]}}
page = post(query, {"order": "ASC", "field": "deadlineDate"})   # one page, so ties in the sort do not matter
print("portal reports", page["totalResults"], "open Horizon Europe records with a deadline from", TODAY)

rows = []
for r in page["results"]:
    m = r["metadata"]
    actions = json.loads(m["actions"][0]) if m.get("actions") else []
    deadlines = sorted({d for a in actions for d in a.get("deadlineDates", [])})
    if deadlines and deadlines[-1] >= TODAY:             # keep calls with a deadline today or later
        rows.append((next(d for d in deadlines if d >= TODAY),
                     m["identifier"][0], r["summary"][:70], m["typesOfAction"][0]))
rows.sort()
for row in rows[:5]:
    print(*row, sep=" | ")
print(len(rows), "of", len(page["results"]), "records on this page still have a deadline")
print("link:", "https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/topic-details/" + rows[0][1])

Output of this code on 11.10.2026 against the live API:

portal reports 63 open Horizon Europe records with a deadline from 2026-10-11
2026-10-13 | HORIZON-CL2-2026-02-TRANSFO-01 | Co-funded European Partnership for Social Transformations and Resilien | HORIZON Programme Cofund Actions
2026-10-14 | ERC-2027-STG | ERC Starting Grant | HORIZON ERC Grants
2026-10-28 | HORIZON-EIC-2026-DEFENCE-01 | EIC defence call | HORIZON EIC Equity Only
2026-10-28 | HORIZON-EIC-2026-PATHFINDERCHALLENGES-01-01 | Advanced Materials for Miniaturised Energy Harvesting Systems | HORIZON EIC Grants
2026-10-28 | HORIZON-EIC-2026-PATHFINDERCHALLENGES-01-02 | Biotechnology for Healthy Ageing | HORIZON EIC Grants
63 of 63 records on this page still have a deadline
link: https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/topic-details/HORIZON-CL2-2026-02-TRANSFO-01

Rules

• Public endpoint only. No login, no cookies, no attempts around blocks.
• At most 2 requests per second, one at a time. Stop on HTTP 401 or 403. On 429 wait the time in the Retry-After header, or stop.
• Send a User-Agent that names your tool and gives your own contact.
• To read a long list, sort by ccm2Id ascending and ask for ccm2Id >= last id seen on page 1 each time. Paging deeper with a sort on a date field repeats and skips records, because many records share a date.
• The portal still lists some old topics as open although every deadline has passed. Check the dates in actions.
• Say what you could not check: the closing time and time zone are on the portal page of the call, and the budget of a topic sits in budgetOverview under the entry whose action starts with the topic identifier.

Limits

About 540 open and 650 forthcoming records exist, and about 18,000 closed ones. A full read of the open and forthcoming list is 13 requests. For scheduled runs, cleaned text, CSV export or the full list in one call, a script or the hosted version is quicker than asking one question at a time.