Vaner
Integrations

MCP Mode

Conceptual overview of Vaner's MCP integration and current tool surface.

Ask ChatGPTAsk Claude

Vaner ships an MCP server so MCP-aware IDEs and agents can pull Prepared Work and evidence-backed context without putting Vaner in the model hot path.

Start the server

vaner mcp --path .

The server speaks stdio by default. Clients invoke vaner mcp --path . as a subprocess and communicate over their own stdin/stdout.

Normal clients should call vaner.prepared_work.dashboard first. It returns a small list of UI-safe cards and hides internal lifecycle, self-evaluation, and raw ranking fields.

When the user inspects a card, call the action returned by that card. Do not infer allowed actions from type alone. Server-authoritative actions are what keep virtual diffs and exports non-mutating by default.

Tools exposed

Prepared Work

ToolPurpose
vaner.prepared_work.dashboardUnified Prepared Work dashboard for UI-safe cards.
vaner.work_products.listDiagnostic list of prepared artifacts.
vaner.work_products.inspectInspect body, evidence, warnings, and allowed actions.
vaner.work_products.exportExport ready content or patch payload without applying it.
vaner.work_products.dismissHide a prepared artifact from normal feeds.
vaner.work_products.feedbackRecord usefulness feedback.

Resolve and navigation

ToolPurpose
vaner.statusHealth and readiness smoke test.
vaner.resolveResolve a prompt into prepared briefing, evidence, and optional draft.
vaner.suggestLikely next prompts or work directions.
vaner.searchSearch Vaner's scenario graph.
vaner.expandDeepen precompute around a scenario or target.
vaner.inspectInspect one scenario/evidence record.
vaner.explainHuman-readable rationale for a resolution.
vaner.feedbackFeedback for resolution/scenario flows.
vaner.warmHint Vaner to precompute around selected targets.
vaner.debug.traceDiagnostic trace data.

Predictions and workspace goals

Predictions remain available for diagnostic and advanced clients, but Prepared Work is the normal user-facing surface.

ToolPurpose
vaner.predictions.activeList ready/drafting predictions.
vaner.predictions.adoptAdopt a specific prediction as the current intent.
vaner.predictions.dashboardDiagnostic prediction dashboard.
vaner.goals.listList workspace goals.
vaner.goals.declareDeclare a new user-authoritative goal.
vaner.goals.update_statusMove a goal between active, paused, abandoned, and achieved.
vaner.goals.deleteDelete a goal.

Setup, policy, sources, artifacts, and Deep-Run

Tool familyPurpose
vaner.setup.*Setup questions, recommendations, apply flow, and setup status.
vaner.policy.showCurrent policy bundle and posture.
vaner.artefacts.*List/inspect/status/influence controls for stored artifacts.
vaner.sources.statusSource health/status.
vaner.deep_run.*Start, stop, inspect, list, and default Deep-Run sessions.

All tools are scoped to the --path Vaner was started with.

Want to wire it up?

When to use MCP vs other modes

  • MCP mode: your IDE orchestrates model calls and needs Vaner-provided Prepared Work/context. Recommended default.
  • Proxy mode: tools that only speak the OpenAI /v1/chat/completions shape.
  • Context-only mode: frameworks that call /v1/context directly.
  • Python SDK: custom scripts.

On this page