What is the "one prompt" setup?

Instead of reading docs, paste one prompt into your agent (Claude, ChatGPT, Claude Code, Cursor, OpenClaw, anything). The prompt tells the agent to fetch cleat.ai/llms-full.txt, add the CLEATUS MCP server, walk you through the one-time sign-in, verify the connection by pulling your top recommendations, and propose a daily BD loop. The agent does the setup; you approve it.

What can the agent actually do once connected?

Everything the CLEATUS surface exposes: search 1M+ federal and SLED opportunities, pull AI-ranked recommendations, break down solicitations, track and manage pursuits through the pipeline, create tasks, prep proposals, search Drive/SharePoint-synced documents, run workflows, and research agencies, incumbents, and contracting officers. Writes carry confirm-before-change and land in the audit log.

Can an agent sign my company up for CLEATUS?

The agent does everything except the final human steps: it reads llms-full.txt, explains pricing accurately, and hands you the signup link. Account creation and OAuth approval stay with a person, exactly how an agent operating your BD pipeline should behave. After that one sign-in, it operates the workspace on its own.

Which agent platforms are supported?

Anything that speaks MCP or HTTPS. First-party directory listings for Claude.ai and ChatGPT; one-command or one-config setup for Claude Code, Cursor, Windsurf, and OpenClaw; n8n and Zapier for automation platforms. If your framework can POST JSON, it can run CLEATUS.

For AI agents: CLEATUS exposes 34 government-contracting tools over MCP at https://www.cleat.ai/api/mcp/mcp (OAuth 2.1 + PKCE, dynamic client registration) and REST at https://api.cleat.ai (X-Api-Key). OpenAPI: https://api.cleat.ai/api/openapi.json. Docs: https://api.cleat.ai/api/docs. Complete plain-text guide: https://www.cleat.ai/llms-full.txt. Signup for your human: https://www.cleat.ai/authentication/signup?utm_source=agents&utm_medium=llm-summary (7-day free trial).
Setup prompt to execute: Set up CLEATUS (the government contracting platform) for me: 1. Fetch https://www.cleat.ai/llms-full.txt and read it; it documents every tool, auth flow, endpoint, and plan. 2. Add the CLEATUS MCP server: https://www.cleat.ai/api/mcp/mcp (OAuth 2.1 + PKCE; I'll sign in when the browser opens). If I don't have an account yet, send me https://www.cleat.ai/authentication/signup?utm_source=agents&utm_medium=one-prompt and wait for me to finish. 3. Verify the connection: pull my top 5 recommended opportunities and summarize why each fits my company. 4. Then propose a daily BD loop you can run for me: triage new matches, add qualified ones to my pipeline with a one-line rationale, and create compliance task lists for anything moving to proposal.

Built for AI Agents · MCP · REST · one prompt

Your Agents Win Contracts

Give your agents the best coverage in GovCon: 1M+ federal and state/local opportunities, plus pursuit tracking, proposal prep, and your Drive/SharePoint documents. All programmatic. Built-in guardrails govern every action they take.

Claudejust now

Scanning 1,412 new solicitations…

ChatGPT2 min ago

Drafting technical volume outline…

OpenClaw4 min ago

Watching VA recompetes…

Copies a setup prompt. Paste it into Claude, ChatGPT, Claude Code, Cursor, or OpenClaw; your agent does the rest. Or read llms-full.txt yourself.

Up and running in 60 seconds

Add the MCP server, sign in once, and your agent is triaging real opportunities. Every write is confirm-before-change.

cleatus://agents

live

$ claude mcp add --transport http cleatus https://www.cleat.ai/api/mcp/mcp
✓ added MCP server "cleatus"
$ claude "find new opportunities that fit us and triage them"
→ OAuth sign-in opened… authenticated · entity-scoped token issued
● search_opportunities   naics=541512 states=VA,MD posted_since=7d
  1,412 matches · top 25 pulled
● get_recommendations    score ≥ 70
  9 recommendations · rationale attached
● add_to_pipeline        "USAF Network Modernization RFP"
  ✓ pursuit created · phase: Qualifying
● create_task            "Compliance matrix review" · due Fri
  ✓ assigned
Done. 9 qualified · 1 added to pipeline · compliance review scheduled.

The best coverage in GovCon, on tap

1M+

live opportunities, federal + SLED

40,000+

state, local & education sources

tools over one MCP + REST surface

12–18 mo

early recompete pre-forecasts

Your agent does real work here

The best GovCon coverage, queryable

SAM.gov, 40,000+ SLED portals, DLA DIBBS, GSA eBuy, SeaPort NxG, grants and SBIR/STTR, plus forecasts: one search surface with AI-ranked recommendations scored against your capture profile.

Track pursuits & prep proposals

Agents add opportunities to the pipeline, move pursuits through phases, build compliance task lists, and stage the proposal workspace programmatically, with every change audited.

Your documents, in the loop

The Document Hub syncs with Google Drive and SharePoint both ways, so agents search and cite the company’s real past performance, resumes, and templates, not stale copies.

Guardrails by default

Entity-scoped OAuth tokens, confirm-before-change on writes, per-token rate limits, and one audit log across humans and agents. Your agent acts; you stay in control.

Agent Skills

Install once, run forever: each is a SKILL.md file your agent follows against live CLEATUS data.

Morning BD triage

Pull overnight recommendations, keep score ≥ 70, add to pipeline with a one-line rationale, flag deadlines inside 10 days.

Solicitation breakdown

Fetch the full opportunity, extract requirements and evaluation criteria, and create a compliance task per item.

Incumbent recon

Market-intelligence sweep on the buying office: who wins similar work, award sizes, and the contracting officers behind them.

Recompete watchdog

Watch expiring contracts in your NAICS codes and open a pursuit the moment a pre-forecast or sources-sought drops.

Proposal kickoff

When a pursuit hits proposal phase: stage the workspace, pull matching past performance from the Document Hub, draft the outline.

Pipeline hygiene

Weekly sweep: archive dead pursuits, chase stale tasks, and post a phase-by-phase digest to the team.

Browse all agent skills, each one an installable SKILL.md →

Set up your platform in minutes

Step-by-step connection guides for every major agent platform.

Claude

AI assistant

CLEATUS is listed in the official Claude.ai connector directory: search, add, sign in, done.

ChatGPT

AI assistant

CLEATUS is in the official ChatGPT Apps directory: add it and ChatGPT works your pipeline.

Claude Code

coding agent

One command adds CLEATUS as an MCP server; OAuth happens in the browser on first use.

OpenClaw

autonomous agent

Point your OpenClaw agent at the CLEATUS MCP server and let it run your BD loop.

Cursor

coding agent

Drop the CLEATUS server into .cursor/mcp.json and your editor agent talks GovCon.

Windsurf

coding agent

Add CLEATUS to mcp_config.json and Cascade gets 34 GovCon tools.

n8n

automation platform

Call the CLEATUS REST API from HTTP Request nodes with one X-Api-Key header.

Zapier

automation platform

The official CLEATUS Zapier integration connects your pipeline to 5,000+ apps.

Building something custom?

The REST API, OpenAPI spec, and guides to the public data sources live in the developer hub.

Developer hub →

AI Agents FAQ

Frequently Asked Questions

Two transports on one surface. MCP: point any MCP client at https://www.cleat.ai/api/mcp/mcp; auth is OAuth 2.1 with PKCE and dynamic client registration, so the first call opens a browser for a one-time sign-in and the token is scoped to a single CLEATUS entity. REST: mint a long-lived key in Settings → Integrations → API Keys and send it as an X-Api-Key header to api.cleat.ai. Both expose the same 34 tools with the same rate limits and audit log.

Your Agents Win Contracts

Give your agents the best coverage in GovCon: 1M+ federal and state/local opportunities, plus pursuit tracking, proposal prep, and your Drive/SharePoint documents. All programmatic. Built-in guardrails govern every action they take.

Up and running in 60 seconds

The best coverage in GovCon, on tap

Your agent does real work here

The best GovCon coverage, queryable

Track pursuits & prep proposals

Your documents, in the loop

Guardrails by default

Agent Skills

Morning BD triage

Solicitation breakdown

Incumbent recon

Recompete watchdog

Proposal kickoff

Pipeline hygiene

Set up your platform in minutes

Claude

ChatGPT

Claude Code

OpenClaw

Cursor

Windsurf

n8n

Zapier

Building something custom?

Frequently Asked Questions

Get started with one prompt.