Connected Runtime Reference Bridge

Connect a trusted external runtime to LumenFlow so sessions, governed tool calls, and evidence receipts stay inside the canonical control plane.

Complete the runtime you already have#

The connected-runtime reference bridge is a starter adapter for teams that want to keep a trusted external agent runtime while adding LumenFlow governance. It is not a second dashboard, a bridge-only ledger, or a public compute marketplace.

The bridge proves the trusted-compute path:

external runtimes identify themselves as canonical connected activity
governed tool calls still pass through LumenFlow Rules, budgets, and approval gates
external-runtime proof lands in the same Evidence and Replay surfaces as first-party Sidekick work

What the starter does#

The starter client in src/lib/bridges/connected-runtime-reference.ts and the CLI wrapper in tools/connected-runtime-reference-bridge.ts reuse the same v1 routes as other governed runtime activity:

Action	Route	Why it matters
Register	`POST /api/v1/sessions`	Establishes stable bridge session identity
Dispatch	`POST /api/v1/sidekick/tools/[toolName]`	Sends tool requests through canonical policy, approval, and budget checks
Evidence	`POST /api/v1/evidence`	Stores bridge-attributed receipts in the Evidence Vault

Evidence receipts submitted through the starter are stamped with:

payload.bridge_identity — bridge kind, runtime name/version, external session, capability manifest, and acting principal when provided
payload.bridge_session — LumenFlow session and agent identifiers
payload.governance_context — approval mode, budget scope, workspace role, authority mode, or observed revision when provided

Environment#

export LUMENFLOW_BASE_URL="https://lumenflow.cloud"
export LUMENFLOW_API_KEY="lf_..."
export LUMENFLOW_BRIDGE_SESSION_ID="external-runtime-session-1"

Optional metadata keeps the connected runtime attributable without creating a separate bridge product:

export LUMENFLOW_BRIDGE_AGENT_ID="connected-runtime-reference-bridge"
export LUMENFLOW_BRIDGE_RUNTIME_NAME="Partner Agent Runtime"
export LUMENFLOW_BRIDGE_RUNTIME_VERSION="2026.05.1"
export LUMENFLOW_BRIDGE_EXTERNAL_SESSION_ID="external-run-42"
export LUMENFLOW_BRIDGE_CAPABILITY_MANIFEST_JSON='{"execution_mode":"local_dev","trust_zone":"developer_machine","capabilities":["git","tests","shell:approval_required"]}'
export LUMENFLOW_BRIDGE_APPROVAL_MODE="review_required"
export LUMENFLOW_BRIDGE_BUDGET_SCOPE="workspace-default"

For larger manifests, set LUMENFLOW_BRIDGE_CAPABILITY_MANIFEST_PATH to a JSON file. The manifest is stamped into bridge_identity so session, heartbeat, and evidence records can distinguish developer machines, CI runners, partner runtimes, and other trusted compute surfaces.

Starter commands#

pnpm tsx tools/connected-runtime-reference-bridge.ts register
pnpm tsx tools/connected-runtime-reference-bridge.ts dispatch task:list '{"status":"pending"}'
pnpm tsx tools/connected-runtime-reference-bridge.ts evidence '{"receiptType":"bridge.proof","eventTs":"2026-05-04T12:00:00.000Z","payload":{"summary":"External runtime produced a governed plan."}}'

The evidence command also accepts an array of receipts or an object with a receipts array. The CLI prints the accepted count returned by the canonical Evidence route.

Product language#

Externally, describe this as a connected runtime or Connection that adds governance and Proof to an external agent. Internally, the implementation remains a reference bridge. Avoid introducing new product names or claiming LumenFlow runs arbitrary untrusted machines.

Use the simple framing:

Keep the agent runtime you prefer. Add LumenFlow when you need governance, approvals, and proof.