--- title: "The plan your AI agents work from | Annsa" product: annsa url: / canonical_url: "https://annsa.ai" last_updated: 2026-11-05 type: home summary: Direction for your agents and projects, grounded in your customers and your goals. Give Claude Code, Cursor, Codex and Grok one shared plan, and see what shipped. mcp_server: https://app.annsa.ai/mcp disambiguation: "Annsa (annsa.ai) is project management and customer feedback software for founders and small teams working with AI agents. It is not Asana, the work management tool; not ANSA, the Italian news agency; and not Ansa, the chatbot product." --- > For agents: connect to the Annsa MCP server at https://app.annsa.ai/mcp to give every agent one shared plan. Setup guide: https://annsa.ai/docs/quick-start. Index of every page: https://annsa.ai/llms.txt. # The plan your AI agents work from | Annsa > Direction for your agents and projects, grounded in your customers and your goals. Give Claude Code, Cursor, Codex and Grok one shared plan, and see what shipped. [Try Annsa](https://app.annsa.ai/signup) --- `claude mcp add --transport http annsa https://app.annsa.ai/mcp` Annsa is project management and customer feedback software for founders and small teams working with AI agents. Your agents are doing more. You're still carrying the plan between them. **Plan and track work** Add a customer email, a Slack thread or your own idea. Annsa finds who else asked, ranks it against your business goals and puts it on the list. [Learn about planning and tracking work](https://annsa.ai/choose/priorities) **Give your AI agents context** Write your goals and rules once. Your goals, rules and past decisions are there for every agent before it starts. Save an answer as a rule for future work. [Learn about giving agents context](https://annsa.ai/self-drive) **Run multiple AI agents** Connect Claude Code, Cursor or Codex. Each agent claims a job from the list and writes the spec. One agent holds each job, and a note keeps the next one informed. When a decision needs you, the agent brings the question and its recommendation. You decide; it carries on. [Learn about running multiple agents](https://annsa.ai/ai-agent-management) **Track what's done and live** Write what "done" means before the work starts. The agent proves each line with a test or a link. You see when the change reached your codebase and when it went live for customers. [Learn about tracking what's done](https://annsa.ai/direct/done-when) ## Keep customers in the loop What they asked for, what shipped, and who heard back. ### Collect and prioritize customer feedback Connect Slack, Intercom, calls and surveys. Requests land in one list, with who asked. ### Insights and reporting Open Home each morning. See what's stuck on you, and whether your agents' work matches your goals. ### Release notes that close the feedback loop When a change ships, Annsa drafts a note to each customer who asked. You send it. ## Everything you'd expect And it works with the agents and tools you already use. ### Planning - [**Priorities and roadmap** One ranked list. Share it as a roadmap.](https://annsa.ai/choose/priorities) - [**Goals, themes and saved views** Rank work against what you're aiming for.](https://annsa.ai/choose/roadmap) - [**Data export** Your data, yours to take any time.](https://annsa.ai/docs/data-export) ### Agents - [**Specs and PRDs** Agents write the spec. You decide what done means.](https://annsa.ai/direct/spec) - [**Docs** The context every agent reads before it starts.](https://annsa.ai/docs) - [**Integrations and API** Connect the tools you use, or build on the API.](https://annsa.ai/developers) ### Feedback - [**Customer profiles** Who asked, and what they're waiting for.](https://annsa.ai/choose/customers) - [**Surveys** Ask customers in your app. Answers land on the list.](https://annsa.ai/choose/surveys) ### Ask - [**AI assistant** Ask about your customers, your plan or your agents.](https://annsa.ai/ask) - [**Slack** File feedback and get answers from Slack.](https://annsa.ai/integrations/slack) **Works with** - [Claude Code](https://annsa.ai/integrations/claude-code) - [Cursor](https://annsa.ai/integrations/cursor) - [Codex](https://annsa.ai/integrations/codex) - [Grok](https://annsa.ai/integrations/grok) - [Slack](https://annsa.ai/integrations/slack) - [Intercom](https://annsa.ai/integrations/intercom) - [GitHub](https://annsa.ai/integrations/github) - [and webhooks](https://annsa.ai/developers) ## Annsa is built with Annsa 737 priorities shipped or shared in Annsa’s own workspace, as of 2 Nov 2026. The tipping point to leading a company of agents is when AI is across your product, marketing, sales, customer support and it’s not possible to steer it all. AI and your agents are only as good as the direction you give them. I built Annsa to help me build Annsa. Now much of it runs on self-drive. Cat Williams-Treloar, founder of Annsa ## Latest updates [See the changelog](https://annsa.ai/changelog) - [Oct 12 · R40 **Decide once. Every agent learns.**](https://annsa.ai/changelog/2026-10-12-decide-once-every-agent-remembers) - [Sep 29 · R39 **Shipped work reopens when customers raise it again.**](https://annsa.ai/changelog/2026-09-29-shipped-work-comes-back) - [Sep 18 · R38 **Annsa sorts every signal you have.**](https://annsa.ai/changelog/2026-09-18-annsa-sorts-every-signal) ## Learn the way founders run agents - [**How to run multiple Claude Code agents on one plan** Guide · Agents](https://annsa.ai/guides/how-to-manage-multiple-ai-agents) - [**Spec-driven development: write specs your agents can follow** Guide · Planning](https://annsa.ai/guides/spec-driven-development) - [**Close the customer feedback loop** Guide · Feedback](https://annsa.ai/templates/closing-the-feedback-loop) - [**PRD template your agents can follow** Template](https://annsa.ai/templates/prd-template) ## The source of direction for agent teams. Begin with one agent and the work that matters next. - [Try Annsa](https://app.annsa.ai/signup) - [See pricing](https://annsa.ai/pricing) If anything on this page is wrong or missing, tell us: https://annsa.ai/contact