Skip to content

This release adds 4 notable features for engineering teams evaluating rollout.

✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

custom-agents custom-tools directed-acyclic-graph llm-agent mcp-client mcp-server
+6 more
mcp-servers mcp-tools multi-agent multi-agent-orchestration multi-agent-systems n8n

Summary

AI summary

Four new deterministic step types (IF/Else, Switch, Extract JSON, Print) added without LLM calls.

Full changelog

v1.3.0 — Deterministic Orchestration Steps

What's New
Four new step types that run without any LLM call — zero latency, zero cost, fully predictable.

  • 🔀 IF / Else
    Binary branching based on a Python condition evaluated against your shared state. Use dot-notation to navigate nested keys (state.result.flag). Missing keys safely evaluate to None. Choose the true path or the false path — no LLM needed.

    if_condition: state.score > 7
    if_true_step_id: step_approve
    if_false_step_id: step_reject

  • 🔀 Switch
    Multi-way routing by value match. Evaluate any expression, convert the result to a string, and match it against named cases. Unmatched values fall through to a configurable default route.

    switch_expression: state.category
    switch_cases:
    "buy" → step_buy
    "cart" → step_cart
    switch_default_step_id: step_general

  • 📦 Extract JSON
    Pulls structured JSON out of raw text — works with plain objects, markdown code fences (json ), and multiple JSON blocks in a single string (stored as an array). Perfect for parsing an LLM's raw output before routing.

  • 🖨️ Print
    Renders a Markdown or plain-text template with {state.key} and {state.key.nested} placeholders resolved from shared state, then stores the result. Compose reports, notifications, and formatted summaries without an LLM call.

What's Changed

  • feat: add support forextract_json, print, if_else and switch control flow nodes to orchestration engine and UI by @naveenraj-17 in https://github.com/naveenraj-17/synapse-ai/pull/226
  • Refactor code structure for improved readability and maintainability by @naveenraj-17 in https://github.com/naveenraj-17/synapse-ai/pull/227
  • Feature/add four more steps orchestration by @naveenraj-17 in https://github.com/naveenraj-17/synapse-ai/pull/228
  • docs: add documentation for deterministic control-flow and data-handling steps to README by @naveenraj-17 in https://github.com/naveenraj-17/synapse-ai/pull/229

Full Changelog: https://github.com/naveenraj-17/synapse-ai/compare/v1.2.0...v1.3.0

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track synapseorch-ai/synapse-ai](https:

Get notified when new releases ship.

Sign up free

About synapseorch-ai/synapse-ai](https:

All releases →

Beta — feedback welcome: [email protected]