Skip to Content
Aionis currently ships as a local-first Lite Runtime for developer machines and self-managed deployments.
Examples & ProofOverview

Examples & Proof

Each result below is generated by a real local Runtime path in this repository. These examples are not benchmark leaderboards. They are contract proofs: they show the product loop running end to end and returning bounded artifacts.

Runnable Examples

For a 3-5 minute external walkthrough, start with the Claude Code Demo.

ExampleRun commandWhat it proves
SDK quickstartnpm run -s runtime:quickstart:sdkSDK can complete remember -> guide -> feedback -> measure -> snapshot.
HTTP quickstartnpm run -s runtime:quickstart:httpRaw product facade works without SDK helpers.
Multi-agent quickstartnpm run -s runtime:quickstart:multi-agentPlanner, worker, verifier, and reviewer can share execution memory.
Memory Firewall quickstartnpm run -s runtime:quickstart:memory-firewallExternal candidates can be governed before prompt use.
Memory Firewall A/B demonpm run -s runtime:e2e:memory-firewall-abUnsafe external candidates are blocked from direct-use context.
Claude Code MCP demonpm run -s runtime:quickstart:claude-code-mcpThe published MCP bridge can record execution evidence, compile context, and replay a decision for Claude Code / Cursor style clients.
Flight Recorder incident demonpm run -s runtime:e2e:flight-recorder-incidentAionis can replay what the Agent could see and why.
Loop Engineering profilenpm run -s runtime:e2e:loop-engineering-profileLoop iterations can be observed, guided, attributed, and measured.

Generated Artifacts

ArtifactContract
SDK quickstart aionis_sdk_quickstart_result_v1
HTTP quickstart aionis_http_quickstart_result_v1
Multi-agent quickstart aionis_multi_agent_quickstart_result_v1
Memory Firewall quickstart aionis_memory_firewall_quickstart_result_v1
Memory Firewall A/B demo aionis_memory_firewall_ab_demo_result_v1
Claude Code MCP demo aionis_claude_code_mcp_demo_result_v1
Claude Code real transcript real MCP connection transcript
Flight Recorder incident demo aionis_flight_recorder_incident_demo_result_v1
Loop Engineering profile aionis_loop_engineering_profile_result_v1

What To Look For

When you inspect the generated JSON, check for these surfaces:

SurfaceWhy it matters
guide_trace_idProves feedback can be tied to the exact guide exposure.
agent_contextShows what the Agent actually receives.
memory_use_receiptShows why memory was used, suppressed, or rehydrated.
memory_decision_traceShows the operator-level decision chain.
effect_reportShows whether Aionis can claim a measured product effect.
flight_recorder_reportShows after-incident replay without raw prompt dumping.

What These Examples Do Not Claim

They do not claim that Aionis solves arbitrary external coding tasks by itself. Aionis is not an autonomous runner. It does not replace the Agent host, shell, browser, CI, or verifier.

They show the part Aionis owns:

memory evidence -> governed context -> attributed outcome -> measured effect

Minimal Agent Source

The shortest complete TypeScript host example is:

docs/examples/minimal-agent.ts 

Use it as the reference shape for your own Agent loop.