Teamwork for AI agents
Teamwork is project management software for planning tasks, tracking time, and coordinating people across client and internal projects. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Teamwork.
Log time against a task
Record time spent on a task so project hours and client billing stay accurate.
List tasks in a project
Pull the current task list for a project as a read-only status check.
Create a task from a request
Add a new task with its details after a request comes in, ready to assign to someone.
12 Teamwork actions, graded by risk.
Every Teamwork action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
8Look things up. Allowed by default.
- teamwork.list_tagsList Tags
- teamwork.list_tasksList Tasks
- teamwork.list_peopleList People
- teamwork.list_projectsList Projects
- teamwork.list_milestonesList Milestones
- teamwork.list_task_listsList Task Lists
- teamwork.search_resourcesSearch Teamwork
- teamwork.list_time_entriesList Time Entries
write
4Create and change things. Allow, or ask the user first.
- teamwork.create_tagCreate Tag
- teamwork.create_taskCreate Task
- teamwork.update_taskUpdate Task
- teamwork.log_time_entryLog Time Entry
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Teamwork in three steps.
- 01Your users connect TeamworkThey add their Teamwork api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “create Tag” can wait for the user to approve.
- 03Any agent can actYour agent calls Teamwork through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('teamwork', { read: 'allow', write: 'ask', // create_tag destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Teamwork connects.
Users add their Teamwork api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Teamwork connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Teamwork from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraphMore project management apps.
Teamwork and Arc0, answered.
Can I use Teamwork with Claude, ChatGPT or Cursor?
Yes. Connect Teamwork to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Teamwork actions you allow.
How do users connect Teamwork?
Users add their Teamwork api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Teamwork actions can my agent take?
12 in total: 8 read, 4 write and 0 destructive, such as “create Tag”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Teamwork?
Yes. Allow read actions and deny writes in the Teamwork policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Teamwork too?
Yes. The same Teamwork connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Teamwork into your agent.
Your users connect Teamwork once, under your brand. Your agent gets 12 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan