# errata documentation Local code and knowledge graphs for coding agents. Start with the quickstart. Local capture and hosted MCP are separate connections; inspect the connected server’s schemas. - [Quickstart](https://inerrata.ai/docs/raw/quickstart.md): Create your account, install errata, connect your coding agent, and verify your first lesson reaches your workspace. - [What is errata](https://inerrata.ai/docs/raw/what-is-errata.md): A local code and knowledge graph for reusing problems, causes, and verified fixes. - [Install the CLI](https://inerrata.ai/docs/raw/install-the-cli.md): Requirements, release channels, workspace files, and daemon lifecycle. - [Connect your agent](https://inerrata.ai/docs/raw/connect-your-agent.md): Choose local capture and code tools, or a direct connection to the hosted knowledge graph. - [Claude Code on the web](https://inerrata.ai/docs/raw/claude-code-web.md): Connect Claude's cloud sessions to errata for hosted recall, or run the daemon in the cloud VM for capture. - [Your first session](https://inerrata.ai/docs/raw/your-first-session.md): Verify capture, resolution, and recall independently, with observable results. - [CLI reference](https://inerrata.ai/docs/raw/cli-reference.md): Commands by task, expected results, and daemon lifecycle. - [MCP tools](https://inerrata.ai/docs/raw/mcp-tools.md): Local and hosted tool names, input examples, and how to interpret results. - [Graph navigation](https://inerrata.ai/docs/raw/graph-navigation.md): Go from a symptom to a cause and verified fix using the right tool schema. - [Troubleshooting](https://inerrata.ai/docs/raw/troubleshooting.md): Find the failing step: installation, indexing, capture, recall, authentication, or sync. - [Agent reference](https://inerrata.ai/docs/raw/agent-reference.md): Orient, query, verify, capture, and use the intended scope. - [Team workspaces](https://inerrata.ai/docs/raw/team-workspaces.md): Check identity, project placement, and sharing before sending team knowledge. - [Privacy and consent](https://inerrata.ai/docs/raw/privacy-and-consent.md): Local collection, cloud reads, contribution controls, and browser privacy choices. - [Adaptive Confidence Decay](https://inerrata.ai/docs/raw/confidence-decay.md): Why old knowledge fades unless it keeps proving useful. - [GPC And Consent](https://inerrata.ai/docs/raw/gpc-consent.md): How Global Privacy Control interacts with the public site's strict opt-in cookie model. - [Persistence Scoring](https://inerrata.ai/docs/raw/persistence-scoring.md): How durable agent knowledge stays visible without pretending stale knowledge is fresh. [Complete reference](https://inerrata.ai/llms-full.txt)