No install? Use cloud.anythingmcp.com directly.
Sign in, install the Harvest in one click, paste the credentials, mint an MCP API key — done. No Docker, no git clone, no local server to run.
TL;DR
Manage Harvest (time tracking, projects, clients, invoices, expenses) from any AI agent. 10 tools, Bearer auth with account-ID header.
💡 Niente installazione? Vai direttamente su cloud.anythingmcp.com. Accedi, clicca Connectors → Harvest, inserisci le credenziali, genera una MCP API key — fatto. Niente Docker, niente
git clone, niente server locale.
Harvest + Gemini
Manage Harvest (time tracking, projects, clients, invoices, expenses) from any AI agent. 10 tools, Bearer auth with account-ID header.
Prerequisiti
Le istruzioni di setup complete sono incluse nel connettore stesso (visibili nello store quando lo selezioni). Le variabili d'ambiente richieste sono:
HARVEST_ACCESS_TOKEN, HARVEST_ACCOUNT_ID
Step 1 — Ottieni le credenziali
This connector wraps the Harvest API v2 (api.harvestapp.com/v2).
Setup:
- Log into https://id.getharvest.com → bottom-left avatar → Developers → Create new personal access token.
- Copy the token and the Account ID shown on the same page (a 7-digit number like
1234567). - Set both:
HARVEST_ACCESS_TOKENandHARVEST_ACCOUNT_ID.
Authentication: every request requires BOTH:
Authorization: Bearer ${HARVEST_ACCESS_TOKEN}Harvest-Account-Id: ${HARVEST_ACCOUNT_ID}- A
User-Agentheader (Harvest API requires it — the adapter setsAnythingMCP (support@anythingmcp.com)).
…(continued in the in-app connector instructions)
Step 2 — Installa l'adapter
git clone https://github.com/HelpCode-ai/anythingmcp.git
cd anythingmcp && docker compose up -d
Step 3 — Aggiungi il connettore in Gemini
Gemini CLI legge i server MCP da ~/.gemini/settings.json (o %APPDATA%\gemini\settings.json su Windows). Aggiungi:
{
"mcpServers": {
"anythingmcp": {
"httpUrl": "https://cloud.anythingmcp.com/mcp",
"headers": { "Authorization": "Bearer YOUR_MCP_API_KEY" }
}
}
}
- Ottieni la tua MCP API key da AnythingMCP → Profilo → MCP API Keys → Nuova Key.
- Salva il file e riavvia
gemini. - Esegui
/mcpnella Gemini CLI —Harvestdovrebbe essere elencato come disponibile. - Vertex AI Studio: passa
https://cloud.anythingmcp.com/mcpnell'arraytoolsdella richiesta con lo stesso header Bearer.
Tool disponibili
| Tool | What it does |
|---|---|
harvest_who_am_i | Return the currently authenticated user: id, first_name, last_name, email, is_admin, default_hourly_rate, weekly_capacity |
harvest_list_clients | List clients |
harvest_list_projects | List projects |
harvest_list_tasks | List task templates (the canonical Harvest 'task' types: 'Programming', 'Design', etc |
harvest_list_time_entries | List time entries |
harvest_create_time_entry | Log time on a project + task |
harvest_update_time_entry | Update fields on an existing time entry (notes, hours, project, task) |
harvest_stop_timer | Stop a currently-running time entry timer |
harvest_list_invoices | List invoices |
harvest_list_expenses | List expenses |
FAQ
Gemini 1.5 Pro o 2.x supportano MCP? Sì — Gemini CLI ≥ 0.4 e la Vertex AI tools API accettano connettori MCP httpUrl con header Bearer.
Prossimi passi
Questa guida ti è stata utile?