# Docs


- **Get started**
- [Overview](/docs/overview): Cards is a local-first VS Code extension that organizes coding-agent activity into durable, Git-backed work directories.
- [Installation](/docs/installation): Install the Cards extension in VS Code and link it to your local agent CLI.
- [Getting started](/docs/getting-started): Create your first card, launch a local agent session, inspect the changes in the attribution tree, and merge the work.
- [Common workflows](/docs/common-workflows): Task-shaped recipes for running parallel cards and resuming interrupted sessions.

- **Account and plans**
- [Pricing and plans](/docs/pricing-and-plans): Two tiers: a free tier that needs no account, with a local 50-card bank that accrues +5 per week; and Pro at $30 per month for unlimited card creation, flat and not metered.

- **Concepts**
- [Cards and content](/docs/cards-and-content): A card is the durable per-card git repository for one coding-agent job, holding the description, plans, comments, and attachments that the work accumulates.
- [Actions and agent modes](/docs/actions-and-agent-modes): An action is a named operation that carries a card's context into a configured harness; the default configuration registers four — launch, interview, chat, and captain.
- [Lifecycle, status, and gates](/docs/lifecycle-status-and-gates): A card holds one of five status values, and its plan and merge gates are advisory checkpoints that prompt for approval without hard-blocking transitions.
- [Attribution and version control](/docs/attribution-and-version-control): How Cards attributes changes to users, agents, or the system, and how the attribution tree projects a card's commits onto your active worktree.
- [Sessions and streams](/docs/sessions-and-streams): How Cards records agent runs, streams real-time console output, and persists transcripts in the card repository.

- **Configure**
- [Configuration, rules, and scopes](/docs/configuration-rules-and-scopes): How Cards loads, merges, and overrides configuration layers, and the complete set of settings you can edit across project and user boundaries.
- [Permissions and safety](/docs/permissions-and-safety): How Cards divides safety responsibility between Git-hook metadata validation and agent harness execution permissions.

- **Reference**
- [Privacy and data](/docs/privacy-and-data): Review our local-first privacy architecture: how card data stays on your local disk, how licensing is managed, and how token traffic flows directly to your model provider.
- [What Cards does not do](/docs/boundaries): Understand the boundaries of the Cards extension: we are not the agent harness, do not run the agent loop, host models, store card content in the cloud, or impose workflow gates beyond the two optional plan and merge approval gates.
- [Troubleshooting and support](/docs/troubleshooting-and-support): Diagnose and resolve common issues: recovering crashed sessions, moving cards to Done, cleaning orphaned worktrees, and fixing registration problems.
- [Glossary and reference](/docs/glossary-and-reference): A comprehensive glossary of Cards terms, followed by a reference of the status values, contributor classes, default actions, and gate fields.