Aircover Developer & Agent Resources

Endpoint

https://api.aircover.ai/mcp

Transport: Streamable HTTP · Auth: OAuth 2.0 (PKCE, dynamic client registration) · Manifest: /.well-known/mcp.json

Available tools

Add to Claude Code

claude mcp add --transport http aircover https://api.aircover.ai/mcp

Add to claude.ai / other MCP clients

Add a custom connector with the URL https://api.aircover.ai/mcp. The client discovers OAuth automatically and walks the user through sign-in. An Aircover account is required; new users receive a trial license on sign-up.

Discovery endpoints

https://api.aircover.ai/.well-known/oauth-authorization-server

https://api.aircover.ai/.well-known/oauth-protected-resource

Both are also reachable from this domain: www.aircover.ai/.well-known/… redirects to the authorization server.

Scopes & permission model

The mcp scope grants access to the MCP tool surface only: read access to meetings, transcripts, AI agent results, deal qualification, reports, teams, and indexed documents. It does not grant account administration, billing, user management, or write access to CRM integrations.

Permissions are layered: a token's scope caps what the client may request, and every request is additionally bounded by the authorizing user's own role and organization, so an agent can never see data the user themselves could not. Tokens are revocable at any time via the revocation endpoint.

Grant types: authorization_code, refresh_token · Response type: code · Token endpoint auth: client_secret_basic, client_secret_post · Scopes are requested at authorization time; request only what you need.

• /llms.txt: full product context, when-to-use guidance, and integration docs for LLM agents.
• /.well-known/mcp.json: MCP server manifest (registry server.json format).
• /sitemap.xml: every page on this site.
• Markdown content negotiation: request any page on this site with Accept: text/markdown and receive clean markdown instead of HTML (acceptmarkdown.com convention).
• Real 404s: nonexistent paths return HTTP 404 with recovery links, never a soft-200.

1 · Discover the authorization server

curl https://api.aircover.ai/.well-known/oauth-authorization-server

2 · Register a client (RFC 7591, no manual key issuance)

curl -X POST https://api.aircover.ai/oauth/register \ -H "Content-Type: application/json" \ -d '{"client_name": "My Agent", "redirect_uris": ["https://example.com/callback"], "grant_types": ["authorization_code", "refresh_token"], "scope": "mcp"}'

3 · Authorize (browser) and exchange the code

curl -X POST https://api.aircover.ai/oauth/token \ -d grant_type=authorization_code -d code=$CODE \ -d redirect_uri=https://example.com/callback \ -d code_verifier=$PKCE_VERIFIER \ -d client_id=$CLIENT_ID -d client_secret=$CLIENT_SECRET

4 · Call the MCP endpoint

curl -X POST https://api.aircover.ai/mcp \ -H "Authorization: Bearer $ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{"jsonrpc": "2.0", "id": 1, "method": "tools/list"}'

Aircover Skills github.com/Aircover/aircover-skills

Installable Skills for Claude, Gemini, and ChatGPT, built on the Aircover MCP server. Each skill turns your live deal data (transcripts, AI agent analysis, calendar, and Salesforce) into a finished, ready-to-share work product: a path-to-quota game plan with ranked deals and close strategies, mutual action plans, call prep briefs, product sentiment reports, and more. Download a .skill file, add it to Claude once, and ask in plain English, with no setup or coding required.

Aircover Pipeline github.com/Aircover/aircover-pipeline

A Python package and CLI for scripting against the Aircover API. It pulls every meeting in a date range and runs a coaching agent template against each one, writing structured JSON output, ideal for batch analysis, custom reporting, and data pipelines. Also usable as a client library (from aircover_client import AircoverClient).

pip install aircover-pipeline aircover-pipeline --list-templates aircover-pipeline --start 2026-01-01 --end 2026-03-31 --template-id <id> --output-dir ./out

https://api.aircover.ai/login_google/?source=developers

https://api.aircover.ai/login_ms/?source=developers