Forms and captchas
Asking a provider for an offer means a contact form, a captcha and a wait. An agent stops at the first of the three.
CONTENTUM.COM · For AI agents
3012 open cybersecurity and IT infrastructure projects, 673 public tenders, 410,206 CVE records and a passive domain check, readable by any agent. Reading needs no account and no API key; sending a request, a provider interest or a domain check needs a key from the CONTENTUM team.
Checking WebMCP …
Asking a provider for an offer means a contact form, a captcha and a wait. An agent stops at the first of the three.
Who needs a SOC, which tender is open, which CVE is being exploited: the answers sit in pages made for reading, not for calling.
Once a form is sent, there is no id and no status. An agent cannot tell its principal what happened.
MCP server · Streamable HTTP
Protocol versions 2026-07-28 (stateless), 2025-11-25, 06-18 and 03-26. Server card.
A2A agent · 1.0 and 0.3
SendMessage, GetTask, ListTasks, CancelTask; message/send and tasks/get for 0.3. Agent card.
WebMCP
In a browser with navigator.modelContext the same tools are registered on this page for the in-browser agent.
JSON API · OpenAPI 3.1
openapi.json · example: /agents/api/projects?service=managed-soc&limit=3
| Tool | What it does | Kind |
|---|---|---|
search_projects | Open cybersecurity and IT infrastructure projects and public tenders by service, country and words. | read |
get_project | One project or tender by id (CT-0001): services, country, signal, regulations, technologies, deadline. | read |
list_services | The 12 service areas with their open project counts. | read |
get_service | One service area: description, typical tasks, demand signal, open projects, provider count. | read |
list_countries | Countries with counts of company projects, tenders and listed providers. | read |
get_country | One country: open projects by service, tenders, providers by service, latest projects. | read |
get_provider_coverage | How many listed providers offer a service in a country (counts only). | read |
get_market_stats | Totals: open projects, tenders, countries, demand per service and regulation, data date. | read |
search_vulnerabilities | CVE records by vendor or year, or the actively exploited ones (CISA KEV). | read |
get_vulnerability | One CVE: description, CVSS, weaknesses, references and whether it is actively exploited. | read |
list_guides | Buyer guides (NIS2, penetration testing, ISO 27001, OWASP …) in a given language. | read |
get_guide | Full text of one guide with sections, questions and sources. | read |
run_domain_check | Passive security check of a domain: TLS, security headers, SPF, DMARC, security.txt. Returns score and report link. | sends data, tighter limit |
check_nis2_scope | Whether an organisation is probably covered by NIS2, from sector and size. | read |
get_platform_info | What CONTENTUM does, what a request needs, contacts, languages and the agent endpoints. | read |
submit_project_request | A company describes the work it needs. Returns a request id; the team replies by e-mail. | sends data, tighter limit |
register_provider_interest | A service provider asks to be listed or registers interest in a project. Returns an id. | sends data, tighter limit |
get_request_status | Status of a request (id + e-mail). | read |
Projects are anonymised: no company or buyer names in any result. Providers appear as counts. Active scans (domain map, Deep Scan) need the owner's permission and are run by the CONTENTUM team, not by anonymous agents.
Call 1 · Find
See comparable demand by service and country, or get_provider_coverage for the number of providers.
Call 2 · Ask
Describe the work. The request id comes back at once; the team answers to the e-mail given.
Call 3 · Follow up
Request id plus e-mail returns the status, so the agent can report back.
Result appears here.
curl -s https://contentum.com/mcp \
-H 'Content-Type: application/json' -H 'Accept: application/json, text/event-stream' \
-H 'MCP-Protocol-Version: 2026-07-28' -H 'Mcp-Method: tools/call' -H 'Mcp-Name: search_projects' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_projects","arguments":{"service":"managed-soc","country":"DE","limit":3},"_meta":{"io.modelcontextprotocol/protocolVersion":"2026-07-28"}}}'{
"mcpServers": {
"contentum": {
"type": "http",
"url": "https://contentum.com/mcp"
}
}
}curl -s https://contentum.com/a2a -H 'Content-Type: application/json' -H 'A2A-Version: 1.0' \
-d '{"jsonrpc":"2.0","id":1,"method":"SendMessage","params":{"message":{"role":"ROLE_USER","messageId":"m1","parts":[{"data":{"skill":"search_vulnerabilities","input":{"exploited_only":true,"limit":3}},"mediaType":"application/json"}]}}}'curl -s -X POST https://contentum.com/agents/api/project-requests -H 'Content-Type: application/json' \
-d '{"company":"Example Clinic GmbH","email":"[email protected]","description":"24/7 monitoring for 400 endpoints, start in Q1","services":["managed-soc"],"country":"DE"}'Point your agent at the MCP endpoint or hand it the guide for language models. Everything else is described there.