# Visdom Docs

## Overview

- [Visdom](https://docs.visdom.virtuslab.com/visdom.md): An AI-native SDLC platform, self-hosted, built by VirtusLab.
- [The thesis](https://docs.visdom.virtuslab.com/the-thesis.md)
- [The concept](https://docs.visdom.virtuslab.com/the-concept.md)
- [Products and packages](https://docs.visdom.virtuslab.com/products-and-packages.md): The products Visdom is made of, the layers they share, and the packages built from them.
- [Logical architecture](https://docs.visdom.virtuslab.com/logical-architecture.md)
- [Data flow: the life of a change](https://docs.visdom.virtuslab.com/data-flow-the-life-of-a-change.md)
- [Integrations](https://docs.visdom.virtuslab.com/integrations.md)
- [Deployment architecture](https://docs.visdom.virtuslab.com/deployment-architecture.md)
- [Security summary](https://docs.visdom.virtuslab.com/security-summary.md)
- [Getting there from here](https://docs.visdom.virtuslab.com/getting-there-from-here.md)
- [Onboarding for new users](https://docs.visdom.virtuslab.com/onboarding-for-new-users.md)
- [Built by VirtusLab](https://docs.visdom.virtuslab.com/built-by-virtuslab.md)

## Visdom Control Plane

- [Visdom Control Plane](https://docs.visdom.virtuslab.com/visdom-control-plane/readme.md): One place for the projects, people, permissions and credentials every Visdom product works with.
- [Quick start](https://docs.visdom.virtuslab.com/visdom-control-plane/getting-started/quick-start.md): The first hour as an administrator: sign in, create a project, add its repositories, give people access.
- [How it works](https://docs.visdom.virtuslab.com/visdom-control-plane/getting-started/how-it-works.md): Workspace, projects, apps and roles; how the other Visdom products ask the Control Plane, and how changes reach them.
- [What's available today](https://docs.visdom.virtuslab.com/visdom-control-plane/getting-started/whats-available.md): What the Control Plane does in the current release, what works with conditions, and what is not built yet.
- [Signing in and finding your way](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/using.md): Signing in to the Control Plane, the suite's module rail and project switcher, and where each settings page lives.
- [Dashboard](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/dashboard.md): Home: one view of what the suite's products are running, what needs attention and what it costs, for the selected project.
- [Projects and repositories](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/projects-and-repositories.md): The project catalogue every Visdom product works on, and the git repositories that belong to each project.
- [Users and access](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/users-and-access.md): Who has signed in, and what each person may do in each Visdom product: sign-in access, workspace roles and project roles.
- [Secrets](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/secrets.md): Encrypted secrets for the whole workspace or one project: who may add and replace them, and who may read their values.
- [Integrations](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/integrations.md): Slack, the LLM each product uses, and an Atlassian site — set once for the workspace, overridden per project.
- [General settings](https://docs.visdom.virtuslab.com/visdom-control-plane/using-the-control-plane/general-settings.md): Your light or dark theme, and the workspace timezone every product's timestamps use.
- [The permission model](https://docs.visdom.virtuslab.com/visdom-control-plane/permissions/permission-model.md): Apps, roles and scopes for the whole Visdom suite: the catalogue, how a decision is made, who is an admin, and how grants behave.
- [Integrating a product](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/integrating.md): What a Visdom product gets from the Control Plane, and what it implements to be managed by it.
- [Sign-in and machine principals](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/sign-in-and-machine-principals.md): How the Control Plane validates tokens, tells people from service accounts, creates users, and what a product's service account may do.
- [Effective permissions](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/effective-permissions.md): GET /permissions/effective: what a person may do, flat for one project or grouped for all, cacheable, and who may ask.
- [Project sync](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/project-sync.md): How a product hears that a project or its repositories changed: the targets, the webhook, delivery and retries, and what to read back.
- [Repositories and credentials](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/repositories-and-credentials.md): The repository lists, the repository token, secret values and the Slack, LLM and Atlassian settings a product reads from the Control Plane.
- [The suite shell and the dashboard](https://docs.visdom.virtuslab.com/visdom-control-plane/integrating-a-product/suite-shell-and-dashboard.md): The shared shell every Visdom UI uses — module rail, selected project, theme, timezone — and the summary a product serves to the Home dashboard.
- [Deployment](https://docs.visdom.virtuslab.com/visdom-control-plane/operations/deployment.md): Running the Control Plane: the container, PostgreSQL, Keycloak, health checks, replicas, and what it talks to.
- [Configuration reference](https://docs.visdom.virtuslab.com/visdom-control-plane/operations/configuration-reference.md): Every environment variable the Control Plane reads, its default, and what it changes.
- [Keycloak](https://docs.visdom.virtuslab.com/visdom-control-plane/operations/keycloak.md): What the Control Plane needs from the Keycloak realm: its clients, the admin role, and the service accounts of the other products.
- [Observability](https://docs.visdom.virtuslab.com/visdom-control-plane/operations/observability.md): Logs, OpenTelemetry traces and metrics, and how a call is followed across the Visdom products.
- [Releases and upgrades](https://docs.visdom.virtuslab.com/visdom-control-plane/operations/releases-and-upgrades.md): How Control Plane releases are made, which image to run, and how to upgrade safely.
- [API reference](https://docs.visdom.virtuslab.com/visdom-control-plane/reference/api-reference.md): Every Control Plane API route, who may call it, the OpenAPI document, and the error format.

## Visdom AI Gateway

- [Visdom AI Gateway](https://docs.visdom.virtuslab.com/visdom-ai-gateway/readme.md): Two gateways, one console: governed access to language models and MCP tools, self-hosted on your Kubernetes.
- [Quick start](https://docs.visdom.virtuslab.com/visdom-ai-gateway/getting-started/quick-start.md): Both gateways and the console on one machine with Docker Compose: demo data, a first model call, the guardrails at work. No model provider needed.
- [Architecture](https://docs.visdom.virtuslab.com/visdom-ai-gateway/getting-started/architecture.md): The components of Visdom AI Gateway, how a model call and a tool call travel through them, and how they are deployed.
- [What's available today](https://docs.visdom.virtuslab.com/visdom-ai-gateway/getting-started/whats-available.md): What Visdom AI Gateway does today, what is partial, and what is not built yet.
- [LLM gateway](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/llm-gateway.md): One address and one key for every approved model, in the OpenAI and the Anthropic API style. What the LLM gateway checks on each call, and in what order.
- [Calling models](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/calling-models.md): Requests to the LLM gateway in the OpenAI and the Anthropic API style, with curl, Python and TypeScript: chat, messages, streamed Responses, embeddings, tool calling and the model list.
- [Model catalogue and providers](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/model-catalogue.md): The model catalogue in Git (deploy/models.json): aliases, versions side by side, fallbacks, prices, providers and their credentials, and the kill switch in the console.
- [Projects, budgets and limits](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/projects-budgets-and-limits.md): Budgets, model allow-lists and rate limits per project, budgets per person, and what a caller sees when a limit is reached.
- [API keys and personal keys](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/api-keys.md): Issuing API keys for people, applications and service accounts; limiting, rotating and revoking them; personal keys; and creating keys through the console API.
- [Guardrails](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/guardrails.md): What the LLM gateway checks in every request and every answer: secrets refused, personal data masked, Presidio for names and addresses, the bank's own rules, and the output guard.
- [Activity, costs and reporting](https://docs.visdom.virtuslab.com/visdom-ai-gateway/llm-gateway/activity-and-costs.md): Every model call with its project, key, model, tokens, cost and outcome; cost by project, model, key, owner and provider; how calls are counted; CSV exports and the read-only reporting API.
- [MCP gateway](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/mcp-gateway.md): The MCP gateway puts your approved tool servers behind one gateway: one endpoint per project, its own tokens, risk classes, approvals and guardrails on every tool call.
- [Tool servers](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/tool-servers.md): Register an approved MCP tool server, store its credentials in the gateway, and keep the list of servers in Git.
- [Tools, risk classes and approvals](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/tools-risk-and-approvals.md): The catalogue of tools, resources and prompts, the per-tool switch, risk classes, and the four-eyes approval a high-risk tool needs before it reaches a virtual server.
- [Virtual servers and client tokens](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/virtual-servers.md): Create one MCP endpoint per project with a chosen subset of tools, issue client tokens that work only there, and connect clients to it.
- [Guardrails, limits and emergency stop](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/guardrails-and-emergency-stop.md): What the MCP gateway checks on every tool call: secrets and bank rules in arguments, personal data in results, rate limits, an optional OPA policy, and the switches up to an emergency stop.
- [MCP activity](https://docs.visdom.virtuslab.com/visdom-ai-gateway/mcp-gateway/activity.md): Tool calls through the MCP gateway, per tool and per call, with CSV export: an operational view, not the audit record.
- [Claude Code](https://docs.visdom.virtuslab.com/visdom-ai-gateway/clients/claude-code.md): Point Claude Code at the LLM gateway with a gateway key and a model alias, add a project's MCP virtual server, and attribute its calls to AI Tracing sessions.
- [Codex](https://docs.visdom.virtuslab.com/visdom-ai-gateway/clients/codex.md): Point Codex at the LLM gateway's Responses API with a custom model provider, add MCP virtual servers, and get AI Tracing attribution with no extra setup.
- [SDKs, IDEs and agents](https://docs.visdom.virtuslab.com/visdom-ai-gateway/clients/other-clients.md): OpenAI and Anthropic SDKs, other OpenAI-compatible tools, Orchestrator agents and MCP clients: what each needs to use the gateway.
- [The gateway in the Visdom platform](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/platform.md): What each Visdom module gives the gateway, and how the gateway runs on its own or integrated with the platform.
- [Sign-in and roles](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/sign-in-and-roles.md): How people and scripts sign in to the console, what viewer, operator and admin may do, and where each person's role comes from.
- [Projects from Control Plane](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/control-plane-projects.md): Where the gateway's projects come from, how Control Plane's sync webhook blocks and unblocks a project's keys, and how to configure both.
- [Records and sessions in AI Tracing](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/ai-tracing.md): What the gateway delivers to AI Tracing, how delivery survives outages, and how Claude Code and Codex calls become AI Tracing sessions with transcripts.
- [Export to a SIEM](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/siem-export.md): Export the gateways' records as OTLP logs to an OpenTelemetry Collector in front of your SIEM: format, attributes, delivery guarantees and setup.
- [Notifications](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/notifications.md): E-mail notifications through your SMTP relay, routed per event and project, and one chat webhook that receives every event.
- [Evidence and the Policies page](https://docs.visdom.virtuslab.com/visdom-ai-gateway/visdom-platform/evidence.md): The Evidence page's signed chain of admin changes, how to verify it locally and with AI Tracing, and what the Policies page shows as enforced or simulated.
- [Install on Kubernetes (Helm)](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/kubernetes-helm.md): Install the LLM gateway, the MCP gateway and the console on your own Kubernetes cluster with one Helm chart: values, secrets, network policies, pod security, upgrades.
- [Evaluate with Docker Compose](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/docker-compose.md): The reference deployment in deploy/compose: services, profiles, overlays, ports, the .env file and image pinning. For evaluation on one host, not for production.
- [Offline and air-gapped](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/offline-and-air-gapped.md): What the gateway would call out to and how each call is switched off, mirroring the images by digest, verifying signatures, and updating a site with no internet access.
- [Network, proxy and egress](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/network-proxy-and-egress.md): Make the gateway the only path from your workloads to models, run the gateways behind a corporate proxy with your own CA, and control which networks the MCP gateway may call.
- [Configuration reference](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/configuration-reference.md): Every environment variable the console reads, the LLM gateway's and MCP gateway's settings the deployment sets, with defaults and the matching Helm values.
- [Releases and upgrades](https://docs.visdom.virtuslab.com/visdom-ai-gateway/operations/upgrades-and-releases.md): How the console and the Helm chart are versioned, built, scanned and signed, and how a site installs a new release.
- [Console API (OpenAPI)](https://docs.visdom.virtuslab.com/visdom-ai-gateway/reference/api-reference.md): The console's HTTP API: what it covers, how people, scripts and reporting tools authenticate, where the OpenAPI document is served, and an example that sets a budget and issues a key.
- [Status codes and headers](https://docs.visdom.virtuslab.com/visdom-ai-gateway/reference/errors-and-headers.md): Status codes, error bodies and headers of the LLM gateway, the MCP gateway and the console API, with their causes.

## Visdom AI Tracing

- [Visdom AI Tracing](https://docs.visdom.virtuslab.com/visdom-tracing/visdom-ai-tracing.md): The flight recorder for AI-assisted development: what the agent did, what it cost, and proof of it.
- [Quick start](https://docs.visdom.virtuslab.com/visdom-tracing/getting-started/quick-start.md): From nothing to your first traced agent session and commit in about ten minutes, against your organisation's server.
- [How it works](https://docs.visdom.virtuslab.com/visdom-tracing/getting-started/how-it-works.md): The three ways activity reaches Visdom AI Tracing, what the server does with it, and where each piece runs.
- [What's available today](https://docs.visdom.virtuslab.com/visdom-tracing/getting-started/whats-available.md): What Visdom AI Tracing does today, what works with conditions, and what is not built yet.
- [Capturing agent work](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/capture.md): How Visdom AI Tracing captures what a coding agent does: the tracevault CLI, the agent's own hooks and two git hooks.
- [Installing the CLI and signing in](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/installing-the-cli-and-signing-in.md): Install the tracevault CLI, sign in through Keycloak with the device grant, or use an API key in CI.
- [Setting up a repository](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/setting-up-a-repository.md): What tracevault init installs in a repository, how the repository gets registered, and how to check the wiring.
- [Supported agents](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/supported-agents.md): Which coding agents Visdom AI Tracing captures, how each one is hooked in, and what each gets: capture, policy injection, file changes.
- [What a session captures](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/what-a-session-captures.md): The events the CLI streams, what the server derives from them (model, tokens, cost, file changes), and how commits are attributed to sessions.
- [Flows, labels and params](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/context-flows-labels-and-params.md): The flow ID, labels and params the CLI stamps on every captured event: three layers, how they merge, and how to edit them.
- [Workspace mode and project attribution](https://docs.visdom.virtuslab.com/visdom-tracing/capturing-agent-work/workspace-mode-and-project-attribution.md): Capture for headless and multi-repo agents: a global install, binding the repository at runtime, and attributing sessions to a project with or without a repository.
- [LLM proxy](https://docs.visdom.virtuslab.com/visdom-tracing/llm-proxy/llm-proxy.md): The LLM proxy: Claude Code, Codex and other tools call Anthropic or OpenAI through Visdom AI Tracing, which forwards the call, meters it and records it.
- [Connecting Claude Code and Codex](https://docs.visdom.virtuslab.com/visdom-tracing/llm-proxy/connecting-claude-code-and-codex.md): Pointing Claude Code and Codex at the LLM proxy: the base URLs, the credential the proxy accepts, and what to set up first.
- [Credentials and routing](https://docs.visdom.virtuslab.com/visdom-tracing/llm-proxy/credentials-and-routing.md): Which upstream key the proxy uses for a call: personal and instance credentials, the default route, per-model rules, model rewrite and concurrency caps.
- [Attribution and sessions](https://docs.visdom.virtuslab.com/visdom-tracing/llm-proxy/attribution-and-sessions.md): How the proxy ties a call to a person, a project, a repository and a session; the headers a tool can send; the missing-context and repo-conflict policies; transcript capture; and every error the proxy
- [Secret blocking](https://docs.visdom.virtuslab.com/visdom-tracing/llm-proxy/secret-blocking.md): Refusing a proxied request that carries a registered secret before it reaches the model, and redacting registered secrets from captured transcripts.
- [Policies and verification](https://docs.visdom.virtuslab.com/visdom-tracing/policies-and-verification/policies.md): Where policies live, which rules the server enforces, and the three points where they reach an agent: its instructions, the pre-push check and CI.
- [Policies](https://docs.visdom.virtuslab.com/visdom-tracing/policies-and-verification/policies-1.md): The policy model: condition types and which of them the server enforces, actions, severity, scope, the verification phase, and how to create policies in the web app or the API.
- [The pre-push check](https://docs.visdom.virtuslab.com/visdom-tracing/policies-and-verification/pre-push-check.md): What tracevault check sends and evaluates on every git push, its output, why it fails closed, and how verification phases narrow a policy to the agent's final checks.
- [Verifying commits in CI](https://docs.visdom.virtuslab.com/visdom-tracing/policies-and-verification/verifying-commits-in-ci.md): tracevault verify and the GitHub Action: which commits pass, why signing must be on, how the repository is resolved, and the server-side merge gate for pull requests.
- [Agent instructions and MCP](https://docs.visdom.virtuslab.com/visdom-tracing/policies-and-verification/agent-instructions-and-mcp.md): How an agent learns the repository's policies before it works: the session hooks per agent, tracevault agent-policies for CLAUDE.md and AGENTS.md, and the tracevault MCP server.
- [Signing in and finding your way](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/web-app.md): Signing in with Visdom, the module rail and project switcher, the sidebar, project scope, and who sees which part of the web app.
- [Dashboard](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/dashboard.md): The dashboard's period, its spend, developer, session and token tiles, session quality, the compliance score and cache savings, and why its totals can differ from Analytics.
- [Traces](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/traces.md): Sessions with their transcript, events, file changes and commits; commits with their attribution; the timeline of tool calls; line-by-line AI attribution; and branches.
- [Analytics, costs and flows](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/analytics-and-costs.md): The analytics pages – tokens, models, authors, AI attribution, sessions, cost, software and AI tools – how they count hook sessions and ledger calls, and flows across sessions and repositories.
- [Chat](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/chat.md): Ask questions about past sessions and commits in plain language: who can read and who can ask, @ mentions, the references behind an answer, which LLM answers, and how sessions are indexed.
- [Projects and repositories](https://docs.visdom.virtuslab.com/visdom-tracing/web-app/projects-and-repositories.md): Projects and the repositories in them: Control Plane projects and local ones, adding and removing repositories, a repository's policies, activity and commits, its connection to GitHub, the verificatio
- [Signing and the chain](https://docs.visdom.virtuslab.com/visdom-tracing/evidence-and-compliance/signing-and-the-chain.md): What Visdom AI Tracing seals once signing is on: commits in a hash chain, sessions with a signature, AI Gateway records in a chain of their own; how to turn it on, get the public key and verify.
- [Compliance and the audit log](https://docs.visdom.virtuslab.com/visdom-tracing/evidence-and-compliance/compliance-and-audit-log.md): The Compliance Dashboard and Verify Now, the Audit Log and what it records, and the compliance settings: what each one does today.
- [AI Tracing in the platform](https://docs.visdom.virtuslab.com/visdom-tracing/visdom-platform/platform.md): What AI Tracing records, how it connects to Keycloak, Control Plane and AI Gateway, and what each connection needs.
- [Sign-in and permissions](https://docs.visdom.virtuslab.com/visdom-tracing/visdom-platform/sign-in-and-permissions.md): How people, CI and other modules sign in to AI Tracing, and what each role and Control Plane grant allows.
- [Control Plane](https://docs.visdom.virtuslab.com/visdom-tracing/visdom-platform/control-plane.md): What AI Tracing takes from Control Plane: people and their grants, projects and repositories, and the LLM for stories and chat.
- [Records from AI Gateway](https://docs.visdom.virtuslab.com/visdom-tracing/visdom-platform/ai-gateway.md): How AI Tracing receives AI Gateway's records and content, seals them, and turns model calls into cost, sessions and transcripts.
- [Deployment](https://docs.visdom.virtuslab.com/visdom-tracing/operations/deployment.md): The two containers, the database, what must be set before the first start, health checks, and what runs in the background.
- [Configuration reference](https://docs.visdom.virtuslab.com/visdom-tracing/operations/configuration-reference.md): Every environment variable the AI Tracing server and web app read, with defaults, what each does, and which ones stop the server when wrong.
- [Administration](https://docs.visdom.virtuslab.com/visdom-tracing/operations/administration.md): The Settings pages administrators use: proxy keys, pricing, the story and chat LLM, Control Plane sync, API keys, and what is set by configuration instead.
- [Pricing](https://docs.visdom.virtuslab.com/visdom-tracing/operations/pricing.md): Where AI Tracing's model prices come from, how a call's cost is worked out, and how to override a price.
- [Observability](https://docs.visdom.virtuslab.com/visdom-tracing/operations/observability.md): The server's logs, OpenTelemetry traces and metrics, and what to watch.
- [Command reference](https://docs.visdom.virtuslab.com/visdom-tracing/reference/command-reference.md): Every tracevault command and flag, the hook commands, the environment variables, and the files the CLI reads and writes.
- [API reference](https://docs.visdom.virtuslab.com/visdom-tracing/reference/api-reference.md): The AI Tracing HTTP API: where its OpenAPI document and Swagger UI are, how to authenticate, and what each group of routes does.

## Visdom Context Fabric

- [Visdom Context Fabric](https://docs.visdom.virtuslab.com/visdom-context-fabric/visdom-context-fabric.md): The organisational knowledge layer for AI coding agents.
- [Context Fabric in the platform](https://docs.visdom.virtuslab.com/visdom-context-fabric/context-fabric-in-the-platform.md): The compiled knowledge layer: aggregate, serve, govern.
- [User Guide](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide.md): What Context Fabric manages, and how the context reaches an agent.
- [Setting up a project](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/setting-up-a-project.md)
- [Architecture module](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/architecture-module.md)
- [Wiki module](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/wiki-module.md)
- [Memory module](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/memory-module.md)
- [Repo insights module](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/repo-insights-module.md)
- [Using Context Fabric with the Orchestrator](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/using-context-fabric-with-the-orchestrator.md)
- [Using Context Fabric stand-alone](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/using-context-fabric-stand-alone.md)
- [AGENTS.md fragments](https://docs.visdom.virtuslab.com/visdom-context-fabric/user-guide/agents.md-fragments.md)

## Visdom Orchestrator

- [Visdom Orchestrator](https://docs.visdom.virtuslab.com/visdom-orchestrator/visdom-orchestrator.md): Coding agents as declarative flows running on Kubernetes.
- [The Orchestrator in the platform](https://docs.visdom.virtuslab.com/visdom-orchestrator/the-orchestrator-in-the-platform.md): The execution engine: definitions, drivers, triggers and oversight.
- [User Guide](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide.md): Definitions, flows and runs. Everything else hangs off those three.
- [Concepts and the object model](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/concepts-and-the-object-model.md)
- [The catalog: install, fork, upgrade](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/the-catalog-install-fork-upgrade.md)
- [Configuring an agent](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/configuring-an-agent.md)
- [Flows: nodes, ports and edges](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/flows-nodes-ports-and-edges.md)
- [What happens during a run](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/what-happens-during-a-run.md)
- [Humans in the loop](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/humans-in-the-loop.md)
- [Governance: cost, policy, egress, identity](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/governance-cost-policy-egress-identity.md): Cost, policy, egress, identity.
- [Adapters, webhooks and schedules](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/adapters-webhooks-and-schedules.md)
- [Deploying and configuring the controller](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/deploying-and-configuring-the-controller.md)
- [Known limitations](https://docs.visdom.virtuslab.com/visdom-orchestrator/user-guide/known-limitations.md)

## Visdom Code Review

- [Visdom Code Review](https://docs.visdom.virtuslab.com/visdom-code-review/visdom-code-review.md): Multi-layered review for teams shipping AI-generated code.
- [The review model](https://docs.visdom.virtuslab.com/visdom-code-review/the-review-model.md)
- [What it catches](https://docs.visdom.virtuslab.com/visdom-code-review/what-it-catches.md)
- [Page](https://docs.visdom.virtuslab.com/visdom-code-review/page.md)

## Visdom Testing

- [Visdom Testing](https://docs.visdom.virtuslab.com/visdom-testing/visdom-testing.md): Validation that holds when the same model wrote the code and the tests.
- [Architecture tests](https://docs.visdom.virtuslab.com/visdom-testing/architecture-tests.md)
- [Property-based tests](https://docs.visdom.virtuslab.com/visdom-testing/property-based-tests.md)
- [Mutation tests](https://docs.visdom.virtuslab.com/visdom-testing/mutation-tests.md)
- [Contract tests](https://docs.visdom.virtuslab.com/visdom-testing/contract-tests.md)
- [Choosing a shape](https://docs.visdom.virtuslab.com/visdom-testing/choosing-a-shape.md)
- [Page](https://docs.visdom.virtuslab.com/visdom-testing/page.md)

## Visdom Security

- [Visdom Security](https://docs.visdom.virtuslab.com/visdom-security/visdom-security.md): Contain the agent while it works, harden what it ships.
- [Containment](https://docs.visdom.virtuslab.com/visdom-security/containment.md)
- [Continuous AppSec](https://docs.visdom.virtuslab.com/visdom-security/continuous-appsec.md)
- [Page](https://docs.visdom.virtuslab.com/visdom-security/page.md)